Sharding Postgres with Basic tier in Hyperscale (Citus), how why & when

Published 04-17-2021 09:39 AM 4,596 Views
Microsoft

One of the big new things that the Hyperscale (Citus) option in the Azure Database for PostgreSQL managed service enables you to do—in addition to being able to scale out Postgres horizontally—is that you can now shard Postgres on a single Hyperscale (Citus) node. With a new Hyperscale (Citus) feature in preview called “Basic tier”, you can start small and start cost-effectively while being ready to scale out horizontally any time.

 

With the new Basic tier feature in Hyperscale (Citus)—in Preview—you can now:

 

  • Try out Hyperscale (Citus) at a much lower price point, starting at $0.27 USD/hour[1]
  • Shard Postgres on a single Hyperscale (Citus) node so your application is “scale-out-ready”
  • Provision your Dev & Test environments in a more cost-effective manner

 

With Basic tier, you can use all the features you’ve come to expect in a standard Hyperscale (Citus) server group. In other words, Basic tier is not limited in any way: with Basic tier, you don’t have to sacrifice any Hyperscale (Citus) capabilities.

 

You can read all about the Citus feature that underpins this new Basic tier feature in my recent open source blog post about sharding Postgres on a single Citus node, which we first rolled out in the new and spectacular Citus 10 open source release.

 

In this post, let’s walk through how to provision Basic tier on Hyperscale (Citus) in the Azure portal. And then, how to scale out your Hyperscale (Citus) node from Basic tier to Standard tier by adding more nodes in the portal.

What’s the Standard tier you ask? Standard tier is the new name for a multi-node Hyperscale (Citus) cluster, to differentiate a multi-node cluster from a Basic tier with just a single Hyperscale (Citus) node.

 

Also in this post, we’ll show you some of the ways it can be useful to shard Postgres on a single Hyperscale (Citus) node—with the new Basic tier. In particular, I like the cost-effective angle, since at $0.27 USD/hour, you can try out Hyperscale (Citus) for ~8 hours and you’ll only pay $2-3 USD.

 

single-blue-postgres-elephant-surrounded-by-lighter-elephants-1920x1080.jpg

 

A "How-to" guide for trying out Basic tier in Hyperscale (Citus)

 

To start using Basic tier in Hyperscale (Citus) in the Azure Database for PostgreSQL managed service, as always, a good place to start is docs.microsoft.com:

 

  • Use the Quickstart docs: Follow these quickstart instructions to provision your Basic tier in Hyperscale (Citus).

 

All the instructions you need to provision are in the Quickstart doc (link above) for provisioning a Basic tier in Hyperscale (Citus). Still, I want to highlight some of the steps in this visual guide, with screenshots from the Azure portal. Starting with:

 

  • Enable Preview Features: As of the date of publication of this blog post, Basic tier is a preview feature in Hyperscale (Citus), so make sure to click Enable Preview Features. If you read this post in the future, you may not need this step.

 

Figure 1: As of the date of publication of this blog post, the Basic tier is a preview feature, so make sure to click Enable Preview Features.Figure 1: As of the date of publication of this blog post, the Basic tier is a preview feature, so make sure to click Enable Preview Features.

 

Once you enable the preview features and get to the Compute + storage screen in the Azure portal’s provisioning workflow, remember to choose Basic tier:

 

  • Choose Basic tier: your choices are Basic tier and Standard tier. In this post, we’ll provision the Basic tier, which only costs ~$200 USD/month—or $0.27 USD/hour. Later in the post, we’ll add worker nodes and seamlessly scale out to Standard tier.

 

Figure 2: Pick Basic tier to provision a single Hyperscale (Citus) node, to start small and be scale-out-ready. Later in the post, we’ll scale out to Standard tier as well.Figure 2: Pick Basic tier to provision a single Hyperscale (Citus) node, to start small and be scale-out-ready. Later in the post, we’ll scale out to Standard tier as well.

 

As you follow the Basic tier quickstart documentation to provision Hyperscale (Citus), be sure to pay special attention to:

 

  • Set your Firewall rules: In the Networking tab, set your firewall rules per the instructions in the docs, by allowing the client IP address(es) that will be allowed to access the database.

Provisioning can take a few minutes. After you’ve provisioned your Hyperscale (Citus) server group with Basic tier, you’ll see a “Your deployment is complete screen.” You can click the blue “Go to resource” button to go to the Basic tier instance you just created in Azure portal. Now you’re ready to follow the steps below to connect to the database.

 

  • Get your connection strings from the Azure Portal: As you can see in Figure 3 below, you can find the Connection strings tab in the Settings column on the left of the Azure portal screen. When you click on Connection strings, you can see the different choices for database connection strings to connect to your freshly-provisioned Basic tier in Hyperscale (Citus).

 

Figure 3: Find the “Connection strings” which we’ll use to connect to the Hyperscale (Citus) server group.Figure 3: Find the “Connection strings” which we’ll use to connect to the Hyperscale (Citus) server group.

 

In this post, I’ll use psql—an interactive terminal front-end for PostgreSQL—to connect to the Hyperscale (Citus) database and start sharding Postgres by creating distributed tables. (Azure Tip: You can also use psql as part of Azure Cloud Shell, too!)

 

  • Update the password in the connection string: In the screenshot above “your_password” is a placeholder in the connection string. You will want to update “your_password”, to use your actual password in the connection string.
  • Use psql: Execute the psql command in the command line and connect to the database.
  • Start sharding your Postgres tables: Next, you can start creating distributed tables on the Hyperscale (Citus) server group.

 

-- Create a table with the usual PostgreSQL syntax
CREATE TABLE users_table (user_id bigserial primary key, age int);

-- Convert the table to a distributed table
SELECT create_distributed_table('users_table', 'user_id');

 

The Hyperscale (Citus) create_distributed_table function divides the Postgres table across 32 shards, but you can still query Postgres as if all the data were in one table. Because the shards are regular Postgres tables, you can still rely on the wealth of Postgres relational database features like transactions, indexes, database constraints, JOINs, and more.

 

As you can see above, in just a few clicks, we were able to provision a Hyperscale (Citus) database using Basic tier—and then create a distributed table on a single Hyperscale (Citus) node.  

Next, let’s explore why you would want to use this newfangled Basic tier thing anyway? How can sharding Postgres on a single node help you?

 

Shard Postgres on a single Hyperscale (Citus) node with Basic tier, to be “scale-out-ready”

 

Sharding Postgres has long been associated with large-scale data sizes. In fact, when most of you think about how Citus shards Postgres, you probably imagine a Hyperscale (Citus) cluster with 2 or 4 worker nodes, or perhaps 20 or 50 or even 100 worker nodes. But with the introduction of Basic tier on Hyperscale (Citus), we can all think about sharding differently.

 

Even if the data sizes are not large, sharding Postgres on a single Hyperscale (Citus) node may provide immediate benefits. By using a distributed data model, you can get:

 

  • query parallelization for multi-shard queries
  • smaller indexes to create/maintain
  • smaller tables to auto-vacuum (in parallel!), and
  • faster bulk data loads

 

For example, after sharding the users_table above with the create_distributed_table function, the following SQL commands will now run in parallel across the Hyperscale (Citus) shards, which can significantly decrease execution times:

 

-- load data, ingestion happens in parallel across shards
INSERT INTO users_table (age)
       SELECT 20 + (random() * 70)::int
       FROM generate_series(0, 100000);

-- this query runs in parallel across all shards
SELECT avg(age) FROM users_table;

-- index created in parallel across all shards
CREATE INDEX user_age ON users_table (age);

 

When your data no longer fits in memory—or if the database server becomes CPU bound—then with Basic tier in Hyperscale (Citus), your data will already be sharded and you’ll be able to easily add more nodes to your cluster to keep your database performant.

 

In other words, with Basic tier in Hyperscale (Citus) you are already prepared to scale out, or what I like to call “scale-out-ready.”

 

Figure 4:  Sharding Postgres on a single node with Basic tier in Hyperscale (Citus)—thereby adopting a distributed data model from the beginning—can make it easy for you to scale out your Postgres database at any time, to any scale. I like to call this being “scale-out-ready” with Hyperscale (Citus) in Azure Database for PostgreSQL.Figure 4: Sharding Postgres on a single node with Basic tier in Hyperscale (Citus)—thereby adopting a distributed data model from the beginning—can make it easy for you to scale out your Postgres database at any time, to any scale. I like to call this being “scale-out-ready” with Hyperscale (Citus) in Azure Database for PostgreSQL.

 

What happens when you need to add worker nodes to your Hyperscale (Citus) Basic tier?

 

When you’re ready to scale, it’s easy to add worker nodes to your Hyperscale (Citus) Basic tier. After all, your data model is already distributed. And you don’t need to migrate, or to upgrade, or to go through any kind of hassle. You simply add more worker nodes in the portal and graduate from the Basic tier to the Standard tier, as we’ll do together below.

 

When you add worker nodes to an Hyperscale (Citus) cluster, there’s no data on them yet. This is because all the shards are still on the old nodes—in this case, with Basic tier, all the data is on the single Hyperscale (Citus) node. So, initially all these new worker nodes will just be doing nothing.

 

This is where the Hyperscale (Citus) shard rebalancing feature comes in. Shard rebalancing ensures that the shards are distributed fairly across all the nodes in the cluster. The Citus shard rebalancer does this by moving shards from one node to another.

 

While a shard is being moved, all read and write queries can continue. In other words, Hyperscale (Citus) enables online rebalancing—or what some people call zero-downtime rebalancing—of the data.

 

Below, we’ll add new nodes to the Hyperscale (Citus) database cluster and rebalance shards across the cluster. We’ll just add 2 worker nodes, but you can scale out your Hyperscale (Citus) server group to as many nodes as you think required.

  • Go to your Settings in the Azure portal: Look for the Compute + storage tab in the Settings column on the left of the Azure portal screen.
  • Choose Standard tier: Choose Standard tier[2] to configure a multi-node cluster.
  • Add worker nodes: Using the slider in the screenshot below, pick any number of worker nodes you need. Remember that with Basic tier, you only have one Coordinator node and you don’t have any worker nodes. In this post, let’s add 2 worker nodes below.
  • Make it so: Click the “Save” button at the bottom of the portal screen.

 

Figure 5: Switch from Basic tier to Standard tier in Hyperscale (Citus) on the Azure Portal’s Compute + storage screen.Figure 5: Switch from Basic tier to Standard tier in Hyperscale (Citus) on the Azure Portal’s Compute + storage screen.

 

Once you’ve added the new Hyperscale (Citus) worker nodes and you’ve graduated your Basic tier to a Standard tier, you are ready to do some shard rebalancing.

  • You decide when to initiate a shard rebalance operation: For maximum control, the choice of when to run the shard rebalancer is left to you. Hyperscale (Citus) does not automatically rebalance when you add new worker nodes.
  • What if you forget to rebalance shards? If you forget to rebalance you will have some empty worker nodes without any data. The good news is, you will see a recommendation to rebalance in the “Shard rebalancer” tab, per the screenshot below.

 

Figure 6: Screenshot of the Azure portal’s Shard rebalancer screen when rebalancing is recommended. In this scenario, you can see that the 2 newly-added worker nodes (blogtest-w0 and blogtest-w1) each have 0 MiB of data—they are empty!Figure 6: Screenshot of the Azure portal’s Shard rebalancer screen when rebalancing is recommended. In this scenario, you can see that the 2 newly-added worker nodes (blogtest-w0 and blogtest-w1) each have 0 MiB of data—they are empty!

 

You can connect back to the coordinator via psql with the same connection string used earlier.

 

As I mentioned earlier, during the rebalance operation, all read and write queries can continue on the Hyperscale (Citus) database. In other words, your production workloads are not affected by the shard rebalancing.

  • Rebalance shards: Once you are ready, rebalance the data with a single command, rebalance_table_shards.

 

-- move shards to new worker node(s)
SELECT rebalance_table_shards();
NOTICE:  Moving shard 102008 from c.blogtest.postgres.database.azure.com:5432 to w1.blogtest.postgres.database.azure.com:5432 ...
NOTICE:  Moving shard 102009 from c.blogtest.postgres.database.azure.com:5432 to w0.blogtest.postgres.database.azure.com:5432 ...
....
NOTICE:  Moving shard 102028 from c.blogtest.postgres.database.azure.com:5432 to w1.blogtest.postgres.database.azure.com:5432 ...

 

During the shard rebalance operation in Hyperscale (Citus), you can monitor its progress. To do that, follow the steps below:

 

  • Monitor the shard rebalancer: Go to the Azure portal. Open the Shard rebalancer page in Server group management. You will see the message that Rebalancing is underway. And you can follow the progress of the rebalance operation in this page.

 

Figure 7: Follow the rebalance progress on the Azure Portal’s Shard Rebalancer screen while the rebalance is in progress.Figure 7: Follow the rebalance progress on the Azure Portal’s Shard Rebalancer screen while the rebalance is in progress.

 

Now that the shards have been distributed across the Hyperscale (Citus) cluster, your application can use the resources on the worker node(s) as well as the coordinator node. Other than the increased compute, memory, and disk available to your application though—from your application’s perspective, nothing has changed.

 

After adding 2 worker nodes to the Hyperscale (Citus) database cluster, and after rebalancing shards across the cluster, your application is still talking to the same Postgres database. Congratulations, you have seamlessly scaled out your Postgres database with Hyperscale (Citus)!

 

Why use Basic tier in Hyperscale (Citus) to shard on a single node, if you don't need to scale out Postgres right now?

 

You may be wondering:

“Why use Hyperscale (Citus) if I don’t need to scale out Postgres right now?”

Well, if you think that your database is not going to grow in the future (and that your database will remain less than ~100GB), then one of the other deployment options in our Azure Database for PostgreSQL managed service—such as Single Server or Flexible Server—is likely to handle your workload quite well. However, if you expect your application (and therefore your Postgres database) to grow over time, then the answer becomes relevant to you.

 

If you start small with Hyperscale (Citus)—by sharding Postgres on a single Hyperscale (Citus) node Basic tier—then as your application grows (in terms of users or activity or features or database size), you’ll be able to easily add nodes and use the zero downtime shard rebalancer on Hyperscale (Citus). Adding more nodes to your Hyperscale (Citus) server group will enable you to fit more data into memory, have higher I/O bandwidth, and provide more compute power (CPU) for your Postgres database. In other words, you will have all the means to keep your database performant for your application even as your application grows.

 

How to prepare your Postgres database to be scale-out-ready?

 

There are a few things you need to take into account to make Hyperscale (Citus) scale-out-ready with the new Basic tier. In scale-out architectures, the data is distributed based on a sharding key (what we call a distribution column in Citus.) The sharding key you choose will influence not only how the data gets distributed, but also what kind of query performance you will get. So a bit of up-front thinking about your query patterns and your data model can go a long way. For example, it is often not performant (or even possible) to enforce UNIQUE constraints on the Postgres columns that do not include distribution keys.

 

If you follow the data modeling best practices of a distributed database when using the Basic tier in Hyperscale (Citus) in Azure Database for PostgreSQL, you can easily scale out your database. The Citus best practices for scaling out your Postgres database seamlessly include:

 

 

If you build your database conforming to the best practices outlined above, the promise of Hyperscale (Citus) is that you’ll be able to scale out your Postgres database to some pretty large database cluster sizes. The key point here is that once you get used to the mindset of scale-out systems, you’ll realize how easy it is to follow the data modeling best practices.

 

Figure 8: Start small and cost-effectively with the new Hyperscale (Citus) “Basic tier” feature (in Preview), and then if and when you need to, you can easily graduate to “Standard tier” by adding more Hyperscale (Citus) nodes.Figure 8: Start small and cost-effectively with the new Hyperscale (Citus) “Basic tier” feature (in Preview), and then if and when you need to, you can easily graduate to “Standard tier” by adding more Hyperscale (Citus) nodes.

 

Using Basic tier for your development & test environments

 

Another interesting scenario for using Basic tier in Hyperscale (Citus) is with your development and test environments. As I mentioned earlier, the Basic tier does not have any limitations compared to Standard tier. So, for many users, Basic tier can be a practical and cost-effective way of provisioning development and testing databases.

 

And there are more Preview features available in Hyperscale (Citus)

 

As I mentioned earlier in the blog, the Basic tier is currently in preview. In fact, there are several super useful Hyperscale (Citus) features in preview, including:

 

  • Basic tier: Run a server group using only a coordinator node and no worker nodes. An economical way to do initial testing and development and handle small production workloads.
  • PostgreSQL 12 and 13: You can now choose among the latest Postgres versions for your Hyperscale (Citus) server group.
  • Citus 10: Installed automatically on server groups running these Preview features and PostgreSQL 13.
  • Columnar storage: Store selected tables' columns (rather than rows) contiguously on disk. Supports on-disk compression. Good for analytic and data warehousing workloads. (You might also find our open source blog post on Citus 10 columnar compression useful.)
  • Read replicas (currently same-region only): Any changes that happen to the primary server group get reflected in its replica, and queries against the replica cause no extra load on the original. Replicas are a useful tool to improve performance for read-only workloads.
  • Managed PgBouncer: A connection pooler that allows many clients to connect to the server group at once, while limiting the number of active connections. Having an inbound PgBouncer satisfies connection requests while keeping the coordinator node running smoothly.

The Basic tier in Hyperscale (Citus) opens the door to new possibilities

 

We are so excited to bring you a preview of the new Basic tier feature (in preview!) in Hyperscale (Citus) on Azure Database for PostgreSQL.

 

And while “opens the door to new possibilities” may sound lofty, it’s true. Basic tier in Hyperscale (Citus) gives you a way to be “scale-out-ready” on day zero in the cloud. If your application is already running on a single-node Postgres, you now have a way to adopt a sharded data model that allows you to scale your Postgres database as much as you need in the future. In essence, as your application grows and you need to scale out, you won’t be faced with any type of database migration challenges.

 

Some useful links & resources when you are ready to try out Citus:

 

  • Getting started: This Getting Started page, created by one of my teammates, has useful resources no matter which way you prefer to learn: reading, watching, or doing. It’s a +1 for getting started with Citus open source as well as with Hyperscale (Citus) in the Azure Database for PostgreSQL managed service.
  • Quickstart documentation for Hyperscale (Citus): These Quickstart docs for provisioning Basic tier in Hyperscale (Citus) are helpful. (And if you don’t yet have an Azure subscription, just create a free Azure account first.)  
  • Citus open source: Check out the Citus open source repo on GitHub to report issues, check out the source code, and learn more. To try Citus open source, you can follow the installation instructions on the GitHub README or you can download Citus packages to try Citus locally. And you can find all the details about sharding Postgres on a single Citus node in my open source blog post, too.
  • Deep dive into Citus internals: If you want to understand the internals of how Citus shards Postgres as an extension, this recent blog post about one of my teammate’s Citus talks at Carnegie Mellon (with video recording by the CMU team) is a great place to start.

 

And if you have any questions about our Hyperscale (Citus) option in our Azure Database for PostgreSQL managed service, my product teammates would love to hear from you. You can always reach out to us over email at AskAzureDBforPostgreSQL. Or follow us on Twitter @AzureDBPostgres. We’d love to know what you think.


Footnotes

  1. In the East US region on Azure, the cost of a Hyperscale (Citus) Basic tier with 2 vCores, 8 GiB total memory, and 128 GiB of storage on the coordinator node is $0.27/hour or ~$200/month.

  2. The minimum Hyperscale (Citus) coordinator size for Standard tier is 4 vCores with 16 GiB memory and 512 GiB disk. So if you have the smallest Basic tier configuration, then switching to Standard tier will increase your Coordinator node size from 2 vCores to 4vCores (or more if you choose) and from 128 GiB storage to 512 GiB storage (or more if you choose.) If your coordinator node needs to change size when moving from Basic to Standard tiers, your coordinator will need to be restarted. Restarting the coordinator requires a brief downtime. You’ll see a warning before the restart happens, so you can confirm that the downtime is OK at that moment. You would avoid the restart if your Basic Tier already had 4 vCores (or 8 vCores) and 512 GiB storage.
%3CLINGO-SUB%20id%3D%22lingo-sub-2275672%22%20slang%3D%22en-US%22%3ESharding%20Postgres%20with%20Basic%20tier%20in%20Hyperscale%20(Citus)%2C%20how%20why%20%26amp%3B%20when%3C%2FLINGO-SUB%3E%3CLINGO-BODY%20id%3D%22lingo-body-2275672%22%20slang%3D%22en-US%22%3E%3CP%3EOne%20of%20the%20big%20new%20things%20that%20the%20%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fhyperscale%2F%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3EHyperscale%20(Citus)%20option%3C%2FA%3E%20in%20the%20Azure%20Database%20for%20PostgreSQL%20managed%20service%20enables%20you%20to%20do%E2%80%94in%20addition%20to%20being%20able%20to%20scale%20out%20Postgres%20horizontally%E2%80%94is%20that%20you%20can%20now%20shard%20Postgres%20on%20a%20single%20Hyperscale%20(Citus)%20node.%20With%20a%20new%20Hyperscale%20(Citus)%20feature%20in%20preview%20called%20%E2%80%9CBasic%20tier%E2%80%9D%2C%20you%20can%20start%20small%20and%20start%20cost-effectively%20while%20being%20ready%20to%20scale%20out%20horizontally%20any%20time.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EWith%20the%20%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fconcepts-hyperscale-tiers%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3Enew%20Basic%20tier%20feature%3C%2FA%3E%20in%20Hyperscale%20(Citus)%E2%80%94in%20Preview%E2%80%94you%20can%20now%3A%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CUL%3E%0A%3CLI%3E%3CSTRONG%3ETry%20out%20Hyperscale%20(Citus)%3C%2FSTRONG%3E%20at%20a%20much%20lower%20price%20point%2C%20starting%20at%20%240.27%20USD%2Fhour%3CSUP%3E%3CA%20href%3D%22%23ftn1%22%20target%3D%22_self%22%20rel%3D%22nofollow%20noopener%20noreferrer%22%20name%3D%22ftnref1%22%3E%5B1%5D%3C%2FA%3E%3C%2FSUP%3E%3C%2FLI%3E%0A%3CLI%3E%3CSTRONG%3EShard%20Postgres%20on%20a%20single%20Hyperscale%20(Citus)%20node%3C%2FSTRONG%3E%20so%20your%20application%20is%20%E2%80%9Cscale-out-ready%E2%80%9D%3C%2FLI%3E%0A%3CLI%3E%3CSTRONG%3EProvision%20your%20Dev%20%26amp%3B%20Test%20environments%3C%2FSTRONG%3E%20in%20a%20more%20cost-effective%20manner%3C%2FLI%3E%0A%3C%2FUL%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EWith%20Basic%20tier%2C%20you%20can%20use%20all%20the%20features%20you%E2%80%99ve%20come%20to%20expect%20in%20a%20standard%20Hyperscale%20(Citus)%20server%20group.%20In%20other%20words%2C%20Basic%20tier%20is%20not%20limited%20in%20any%20way%3A%20with%20Basic%20tier%2C%20you%20don%E2%80%99t%20have%20to%20sacrifice%20any%20Hyperscale%20(Citus)%20capabilities.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EYou%20can%20read%20all%20about%20the%20Citus%20feature%20that%20underpins%20this%20new%20Basic%20tier%20feature%20in%20my%20recent%20open%20source%20blog%20post%20about%20%3CA%20href%3D%22https%3A%2F%2Fwww.citusdata.com%2Fblog%2F2021%2F03%2F20%2Fsharding-postgres-on-a-single-citus-node%2F%22%20target%3D%22_blank%22%20rel%3D%22noopener%20nofollow%20noreferrer%22%3Esharding%20Postgres%20on%20a%20single%20Citus%20node%3C%2FA%3E%2C%20which%20we%20first%20rolled%20out%20in%20the%20new%20and%20spectacular%20%3CA%20href%3D%22https%3A%2F%2Fwww.citusdata.com%2Fblog%2F2021%2F03%2F05%2Fcitus-10-release-open-source-rebalancer-and-columnar-for-postgres%2F%22%20target%3D%22_blank%22%20rel%3D%22noopener%20nofollow%20noreferrer%22%3ECitus%2010%20open%20source%20release%3C%2FA%3E.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EIn%20this%20post%2C%20let%E2%80%99s%20walk%20through%20how%20to%20provision%20Basic%20tier%20on%20Hyperscale%20(Citus)%20in%20the%20Azure%20portal.%20And%20then%2C%20how%20to%20scale%20out%20your%20Hyperscale%20(Citus)%20node%20from%20Basic%20tier%20to%20Standard%20tier%20by%20adding%20more%20nodes%20in%20the%20portal.%3CBR%20%2F%3E%3CBR%20%2F%3EWhat%E2%80%99s%20the%20Standard%20tier%20you%20ask%3F%20Standard%20tier%20is%20the%20new%20name%20for%20a%20multi-node%20Hyperscale%20(Citus)%20cluster%2C%20to%20differentiate%20a%20multi-node%20cluster%20from%20a%20Basic%20tier%20with%20just%20a%20single%20Hyperscale%20(Citus)%20node.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EAlso%20in%20this%20post%2C%20we%E2%80%99ll%20show%20you%20some%20of%20the%20ways%20it%20can%20be%20useful%20to%20shard%20Postgres%20on%20a%20single%20Hyperscale%20(Citus)%20node%E2%80%94with%20the%20new%20Basic%20tier.%20In%20particular%2C%20I%20like%20the%20cost-effective%20angle%2C%20since%20at%20%240.27%20USD%2Fhour%2C%20you%20can%20try%20out%20Hyperscale%20(Citus)%20for%20~8%20hours%20and%20you%E2%80%99ll%20only%20pay%20%242-3%20USD.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3E%3CSPAN%20class%3D%22lia-inline-image-display-wrapper%20lia-image-align-inline%22%20image-alt%3D%22single-blue-postgres-elephant-surrounded-by-lighter-elephants-1920x1080.jpg%22%20style%3D%22width%3A%20400px%3B%22%3E%3CIMG%20src%3D%22https%3A%2F%2Ftechcommunity.microsoft.com%2Ft5%2Fimage%2Fserverpage%2Fimage-id%2F273453i6EB06A1C3276EF2E%2Fimage-size%2Fmedium%3Fv%3Dv2%26amp%3Bpx%3D400%22%20role%3D%22button%22%20title%3D%22single-blue-postgres-elephant-surrounded-by-lighter-elephants-1920x1080.jpg%22%20alt%3D%22single-blue-postgres-elephant-surrounded-by-lighter-elephants-1920x1080.jpg%22%20%2F%3E%3C%2FSPAN%3E%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CH2%20id%3D%22toc-hId--493713965%22%20id%3D%22toc-hId--493713965%22%3EA%20%22How-to%22%20guide%20for%20trying%20out%20Basic%20tier%20in%20Hyperscale%20(Citus)%3C%2FH2%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3ETo%20start%20using%20Basic%20tier%20in%20Hyperscale%20(Citus)%20in%20the%20Azure%20Database%20for%20PostgreSQL%20managed%20service%2C%20as%20always%2C%20a%20good%20place%20to%20start%20is%20docs.microsoft.com%3A%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CUL%3E%0A%3CLI%3E%3CSTRONG%3EUse%20the%20Quickstart%20docs%3C%2FSTRONG%3E%3A%20Follow%20these%20%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fquickstart-create-hyperscale-basic-tier%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3Equickstart%20instructions%3C%2FA%3E%20to%20provision%20your%20Basic%20tier%20in%20Hyperscale%20(Citus).%3C%2FLI%3E%0A%3C%2FUL%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EAll%20the%20instructions%20you%20need%20to%20provision%20are%20in%20the%20Quickstart%20doc%20(link%20above)%20for%20provisioning%20a%20Basic%20tier%20in%20Hyperscale%20(Citus).%20Still%2C%20I%20want%20to%20highlight%20some%20of%20the%20steps%20in%20this%20visual%20guide%2C%20with%20screenshots%20from%20the%20Azure%20portal.%20Starting%20with%3A%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CUL%3E%0A%3CLI%3E%3CSTRONG%3EEnable%20Preview%20Features%3C%2FSTRONG%3E%3CEM%3E%3A%20%3C%2FEM%3EAs%20of%20the%20date%20of%20publication%20of%20this%20blog%20post%2C%20Basic%20tier%20is%20a%20%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fhyperscale-preview-features%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3Epreview%20feature%3C%2FA%3E%20in%20Hyperscale%20(Citus)%2C%20so%20make%20sure%20to%20click%26nbsp%3B%3CEM%3EEnable%20Preview%20Features%3C%2FEM%3E.%20If%20you%20read%20this%20post%20in%20the%20future%2C%20you%20may%20not%20need%20this%20step.%3C%2FLI%3E%0A%3C%2FUL%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3E%3CSPAN%20class%3D%22lia-inline-image-display-wrapper%20lia-image-align-inline%22%20image-alt%3D%22figure-1-screenshot-azure-portal-hyperscale-citus-enable-preview-features.png%22%20style%3D%22width%3A%20999px%3B%22%3E%3CIMG%20src%3D%22https%3A%2F%2Ftechcommunity.microsoft.com%2Ft5%2Fimage%2Fserverpage%2Fimage-id%2F273480i1172CC48FF17EF11%2Fimage-size%2Flarge%3Fv%3Dv2%26amp%3Bpx%3D999%22%20role%3D%22button%22%20title%3D%22figure-1-screenshot-azure-portal-hyperscale-citus-enable-preview-features.png%22%20alt%3D%22Figure%201%3A%20As%20of%20the%20date%20of%20publication%20of%20this%20blog%20post%2C%20the%20Basic%20tier%20is%20a%20preview%20feature%2C%20so%20make%20sure%20to%20click%20Enable%20Preview%20Features.%22%20%2F%3E%3CSPAN%20class%3D%22lia-inline-image-caption%22%20onclick%3D%22event.preventDefault()%3B%22%3EFigure%201%3A%20As%20of%20the%20date%20of%20publication%20of%20this%20blog%20post%2C%20the%20Basic%20tier%20is%20a%20preview%20feature%2C%20so%20make%20sure%20to%20click%20Enable%20Preview%20Features.%3C%2FSPAN%3E%3C%2FSPAN%3E%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EOnce%20you%20enable%20the%20preview%20features%20and%20get%20to%20the%20%3CSTRONG%3ECompute%20%2B%20storage%3C%2FSTRONG%3E%20screen%20in%20the%20Azure%20portal%E2%80%99s%20provisioning%20workflow%2C%20remember%20to%20choose%20Basic%20tier%3A%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CUL%3E%0A%3CLI%3E%3CSTRONG%3EChoose%20Basic%20tier%3C%2FSTRONG%3E%3A%20your%20choices%20are%20Basic%20tier%20and%20Standard%20tier.%20In%20this%20post%2C%20we%E2%80%99ll%20provision%20the%20Basic%20tier%2C%20which%20only%20costs%20~%24200%20USD%2Fmonth%E2%80%94or%20%240.27%20USD%2Fhour.%20Later%20in%20the%20post%2C%20we%E2%80%99ll%20add%20worker%20nodes%20and%20seamlessly%20scale%20out%20to%20Standard%20tier.%3C%2FLI%3E%0A%3C%2FUL%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3E%3CSPAN%20class%3D%22lia-inline-image-display-wrapper%20lia-image-align-inline%22%20image-alt%3D%22figure-2-screenshot-azure-portal-compute-storage-basic-tier.png%22%20style%3D%22width%3A%20999px%3B%22%3E%3CIMG%20src%3D%22https%3A%2F%2Ftechcommunity.microsoft.com%2Ft5%2Fimage%2Fserverpage%2Fimage-id%2F273481iD32AB2C0DA21903E%2Fimage-size%2Flarge%3Fv%3Dv2%26amp%3Bpx%3D999%22%20role%3D%22button%22%20title%3D%22figure-2-screenshot-azure-portal-compute-storage-basic-tier.png%22%20alt%3D%22Figure%202%3A%20Pick%20Basic%20tier%20to%20provision%20a%20single%20Hyperscale%20(Citus)%20node%2C%20to%20start%20small%20and%20be%20scale-out-ready.%20Later%20in%20the%20post%2C%20we%E2%80%99ll%20scale%20out%20to%20Standard%20tier%20as%20well.%22%20%2F%3E%3CSPAN%20class%3D%22lia-inline-image-caption%22%20onclick%3D%22event.preventDefault()%3B%22%3EFigure%202%3A%20Pick%20Basic%20tier%20to%20provision%20a%20single%20Hyperscale%20(Citus)%20node%2C%20to%20start%20small%20and%20be%20scale-out-ready.%20Later%20in%20the%20post%2C%20we%E2%80%99ll%20scale%20out%20to%20Standard%20tier%20as%20well.%3C%2FSPAN%3E%3C%2FSPAN%3E%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EAs%20you%20follow%20the%20%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fquickstart-create-hyperscale-basic-tier%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3EBasic%20tier%20quickstart%3C%2FA%3E%20documentation%20to%20provision%20Hyperscale%20(Citus)%2C%20be%20sure%20to%20pay%20special%20attention%20to%3A%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CUL%3E%0A%3CLI%3E%3CSTRONG%3ESet%20your%20Firewall%20rules%3C%2FSTRONG%3E%3A%20In%20the%20%3CSTRONG%3ENetworking%20%3C%2FSTRONG%3Etab%2C%20set%20your%20firewall%20rules%20per%20the%20instructions%20in%20the%20docs%2C%20by%20allowing%20the%20client%20IP%20address(es)%20that%20will%20be%20allowed%20to%20access%20the%20database.%3CBR%20%2F%3E%3CBR%20%2F%3E%3C%2FLI%3E%0A%3C%2FUL%3E%0A%3CP%3EProvisioning%20can%20take%20a%20few%20minutes.%20After%20you%E2%80%99ve%20provisioned%20your%20Hyperscale%20(Citus)%20server%20group%20with%20Basic%20tier%2C%20you%E2%80%99ll%20see%20a%20%E2%80%9CYour%20deployment%20is%20complete%20screen.%E2%80%9D%20You%20can%20click%20the%20blue%20%3CSTRONG%3E%E2%80%9CGo%20to%20resource%E2%80%9D%3C%2FSTRONG%3E%20button%20to%20go%20to%20the%20Basic%20tier%20instance%20you%20just%20created%20in%20Azure%20portal.%20Now%20you%E2%80%99re%20ready%20to%20follow%20the%20steps%20below%20to%20connect%20to%20the%20database.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CUL%3E%0A%3CLI%3E%3CSTRONG%3EGet%20your%20connection%20strings%20from%20the%20Azure%20Portal%3C%2FSTRONG%3E%3A%20As%20you%20can%20see%20in%20Figure%203%20below%2C%20you%20can%20find%20the%20%3CSTRONG%3EConnection%20strings%20%3C%2FSTRONG%3Etab%20in%20the%20Settings%20column%20on%20the%20left%20of%20the%20Azure%20portal%20screen.%20When%20you%20click%20on%20Connection%20strings%2C%20you%20can%20see%20the%20different%20choices%20for%20database%20connection%20strings%20to%20connect%20to%20your%20freshly-provisioned%20Basic%20tier%20in%20Hyperscale%20(Citus).%3C%2FLI%3E%0A%3C%2FUL%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3E%3CSPAN%20class%3D%22lia-inline-image-display-wrapper%20lia-image-align-inline%22%20image-alt%3D%22figure-3-screenshot-azure-portal-connection-strings-postgresql.png%22%20style%3D%22width%3A%20999px%3B%22%3E%3CIMG%20src%3D%22https%3A%2F%2Ftechcommunity.microsoft.com%2Ft5%2Fimage%2Fserverpage%2Fimage-id%2F273499iAD77B2C620B3B618%2Fimage-size%2Flarge%3Fv%3Dv2%26amp%3Bpx%3D999%22%20role%3D%22button%22%20title%3D%22figure-3-screenshot-azure-portal-connection-strings-postgresql.png%22%20alt%3D%22Figure%203%3A%20Find%20the%20%E2%80%9CConnection%20strings%E2%80%9D%20which%20we%E2%80%99ll%20use%20to%20connect%20to%20the%20Hyperscale%20(Citus)%20server%20group.%22%20%2F%3E%3CSPAN%20class%3D%22lia-inline-image-caption%22%20onclick%3D%22event.preventDefault()%3B%22%3EFigure%203%3A%20Find%20the%20%E2%80%9CConnection%20strings%E2%80%9D%20which%20we%E2%80%99ll%20use%20to%20connect%20to%20the%20Hyperscale%20(Citus)%20server%20group.%3C%2FSPAN%3E%3C%2FSPAN%3E%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EIn%20this%20post%2C%20I%E2%80%99ll%20use%20%3CA%20href%3D%22https%3A%2F%2Fwww.postgresql.org%2Fdocs%2F13%2Fapp-psql.html%22%20target%3D%22_blank%22%20rel%3D%22noopener%20nofollow%20noreferrer%22%3Epsql%3C%2FA%3E%E2%80%94an%20interactive%20terminal%20front-end%20for%20PostgreSQL%E2%80%94to%20connect%20to%20the%20Hyperscale%20(Citus)%20database%20and%20start%20sharding%20Postgres%20by%20creating%20distributed%20tables.%20(%3CSTRONG%3EAzure%3C%2FSTRONG%3E%20%3CSTRONG%3ETip%3C%2FSTRONG%3E%3A%20You%20can%20also%20use%3CCODE%20style%3D%22color%3A%20black%3B%22%3Epsql%3C%2FCODE%3Eas%20part%20of%20%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fcloud-shell%2Foverview%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3EAzure%20Cloud%20Shell%3C%2FA%3E%2C%20too!)%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CUL%3E%0A%3CLI%3E%3CSTRONG%3EUpdate%20the%20password%20in%20the%20connection%20string%3C%2FSTRONG%3E%3A%20In%20the%20screenshot%20above%20%E2%80%9Cyour_password%E2%80%9D%20is%20a%20placeholder%20in%20the%20connection%20string.%20You%20will%20want%20to%20update%20%E2%80%9Cyour_password%E2%80%9D%2C%20to%20use%20your%20actual%20password%20in%20the%20connection%20string.%3C%2FLI%3E%0A%3CLI%3E%3CSTRONG%3EUse%20psql%3C%2FSTRONG%3E%3A%20Execute%20the%3CCODE%20style%3D%22color%3A%20black%3B%22%3Epsql%3C%2FCODE%3Ecommand%20in%20the%20command%20line%20and%20connect%20to%20the%20database.%3C%2FLI%3E%0A%3CLI%3E%3CSTRONG%3EStart%20sharding%20your%20Postgres%20tables%3C%2FSTRONG%3E%3A%20Next%2C%20you%20can%20start%20creating%20distributed%20tables%20on%20the%20Hyperscale%20(Citus)%20server%20group.%3C%2FLI%3E%0A%3C%2FUL%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CPRE%20class%3D%22lia-code-sample%20language-sql%22%3E%3CCODE%3E--%20Create%20a%20table%20with%20the%20usual%20PostgreSQL%20syntax%0ACREATE%20TABLE%20users_table%20(user_id%20bigserial%20primary%20key%2C%20age%20int)%3B%0A%0A--%20Convert%20the%20table%20to%20a%20distributed%20table%0ASELECT%20create_distributed_table('users_table'%2C%20'user_id')%3B%3C%2FCODE%3E%3C%2FPRE%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EThe%20Hyperscale%20(Citus)%20%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fhowto-hyperscale-modify-distributed-tables%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3Ecreate_distributed_table%3C%2FA%3E%20function%20divides%20the%20Postgres%20table%20across%2032%20shards%2C%20but%20you%20can%20still%20query%20Postgres%20as%20if%20all%20the%20data%20were%20in%20one%20table.%20Because%20the%20shards%20are%20regular%20Postgres%20tables%2C%20you%20can%20still%20rely%20on%20the%20wealth%20of%20Postgres%20relational%20database%20features%20like%20transactions%2C%20indexes%2C%20database%20constraints%2C%20JOINs%2C%20and%20more.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EAs%20you%20can%20see%20above%2C%20in%20just%20a%20few%20clicks%2C%20we%20were%20able%20to%20provision%20a%20Hyperscale%20(Citus)%20database%20using%20Basic%20tier%E2%80%94and%20then%20create%20a%20distributed%20table%20on%20a%20single%20Hyperscale%20(Citus)%20node.%20%26nbsp%3B%3CBR%20%2F%3E%3CBR%20%2F%3ENext%2C%20let%E2%80%99s%20explore%20why%20you%20would%20want%20to%20use%20this%20newfangled%20Basic%20tier%20thing%20anyway%3F%20How%20can%20sharding%20Postgres%20on%20a%20single%20node%20help%20you%3F%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CH2%20id%3D%22toc-hId-1993798868%22%20id%3D%22toc-hId-1993798868%22%3EShard%20Postgres%20on%20a%20single%20Hyperscale%20(Citus)%20node%20with%20Basic%20tier%2C%20to%20be%20%E2%80%9Cscale-out-ready%E2%80%9D%3C%2FH2%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3ESharding%20Postgres%20has%20long%20been%20associated%20with%20large-scale%20data%20sizes.%20In%20fact%2C%20when%20most%20of%20you%20think%20about%20how%20Citus%20shards%20Postgres%2C%20you%20probably%20imagine%20a%20Hyperscale%20(Citus)%20cluster%20with%202%20or%204%20worker%20nodes%2C%20or%20perhaps%2020%20or%2050%20or%20even%20100%20worker%20nodes.%20But%20with%20the%20introduction%20of%20Basic%20tier%20on%20Hyperscale%20(Citus)%2C%20we%20can%20all%20think%20about%20sharding%20differently.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EEven%20if%20the%20data%20sizes%20are%20not%20large%2C%20sharding%20Postgres%20on%20a%20single%20Hyperscale%20(Citus)%20node%20may%20provide%20immediate%20benefits.%20By%20using%20a%20distributed%20data%20model%2C%20you%20can%20get%3A%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CUL%3E%0A%3CLI%3E%3CA%20href%3D%22https%3A%2F%2Ftechcommunity.microsoft.com%2Fdocs.citusdata.com%2Fen%2Fstable%2Fdevelop%2Freference_processing.html%23query-processing%22%20target%3D%22_blank%22%3Equery%20parallelization%3C%2FA%3E%20for%20multi-shard%20queries%3C%2FLI%3E%0A%3CLI%3Esmaller%20indexes%20to%20create%2Fmaintain%3C%2FLI%3E%0A%3CLI%3Esmaller%20tables%20to%20auto-vacuum%20(in%20parallel!)%2C%20and%3C%2FLI%3E%0A%3CLI%3Efaster%20bulk%20data%20loads%3C%2FLI%3E%0A%3C%2FUL%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EFor%20example%2C%20after%20sharding%20the%3CCODE%20style%3D%22color%3A%20black%3B%22%3Eusers_table%3C%2FCODE%3Eabove%20with%20the%3CCODE%20style%3D%22color%3A%20black%3B%22%3Ecreate_distributed_table%3C%2FCODE%3Efunction%2C%20the%20following%20SQL%20commands%20will%20now%20run%20in%20parallel%20across%20the%20Hyperscale%20(Citus)%20shards%2C%20which%20can%20significantly%20decrease%20execution%20times%3A%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CPRE%20class%3D%22lia-code-sample%20language-sql%22%3E%3CCODE%3E--%20load%20data%2C%20ingestion%20happens%20in%20parallel%20across%20shards%0AINSERT%20INTO%20users_table%20(age)%0A%20%20%20%20%20%20%20SELECT%2020%20%2B%20(random()%20*%2070)%3A%3Aint%0A%20%20%20%20%20%20%20FROM%20generate_series(0%2C%20100000)%3B%0A%0A--%20this%20query%20runs%20in%20parallel%20across%20all%20shards%0ASELECT%20avg(age)%20FROM%20users_table%3B%0A%0A--%20index%20created%20in%20parallel%20across%20all%20shards%0ACREATE%20INDEX%20user_age%20ON%20users_table%20(age)%3B%3C%2FCODE%3E%3C%2FPRE%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EWhen%20your%20data%20no%20longer%20fits%20in%20memory%E2%80%94or%20if%20the%20database%20server%20becomes%20CPU%20bound%E2%80%94then%20with%20Basic%20tier%20in%20Hyperscale%20(Citus)%2C%20your%20data%20will%20already%20be%20sharded%20and%20you%E2%80%99ll%20be%20able%20to%20easily%20add%20more%20nodes%20to%20your%20cluster%20to%20keep%20your%20database%20performant.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EIn%20other%20words%2C%20with%20Basic%20tier%20in%20Hyperscale%20(Citus)%20you%20are%20already%20prepared%20to%20scale%20out%2C%20or%20what%20I%20like%20to%20call%20%E2%80%9C%3CSTRONG%3Escale-out-ready%3C%2FSTRONG%3E.%E2%80%9D%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3E%3CSPAN%20class%3D%22lia-inline-image-display-wrapper%20lia-image-align-inline%22%20image-alt%3D%22figure-4-diagram-sharding-postgres-single-node-basic-tier-or-distributed-hyperscale-citus-cluster-transparent.png%22%20style%3D%22width%3A%20999px%3B%22%3E%3CIMG%20src%3D%22https%3A%2F%2Ftechcommunity.microsoft.com%2Ft5%2Fimage%2Fserverpage%2Fimage-id%2F273500i29A7D336E550BD49%2Fimage-size%2Flarge%3Fv%3Dv2%26amp%3Bpx%3D999%22%20role%3D%22button%22%20title%3D%22figure-4-diagram-sharding-postgres-single-node-basic-tier-or-distributed-hyperscale-citus-cluster-transparent.png%22%20alt%3D%22Figure%204%3A%20Sharding%20Postgres%20on%20a%20single%20node%20with%20Basic%20tier%20in%20Hyperscale%20(Citus)%E2%80%94thereby%20adopting%20a%20distributed%20data%20model%20from%20the%20beginning%E2%80%94can%20make%20it%20easy%20for%20you%20to%20scale%20out%20your%20Postgres%20database%20at%20any%20time%2C%20to%20any%20scale.%20I%20like%20to%20call%20this%20being%20%E2%80%9Cscale-out-ready%E2%80%9D%20with%20Hyperscale%20(Citus)%20in%20Azure%20Database%20for%20PostgreSQL.%22%20%2F%3E%3CSPAN%20class%3D%22lia-inline-image-caption%22%20onclick%3D%22event.preventDefault()%3B%22%3EFigure%204%3A%20Sharding%20Postgres%20on%20a%20single%20node%20with%20Basic%20tier%20in%20Hyperscale%20(Citus)%E2%80%94thereby%20adopting%20a%20distributed%20data%20model%20from%20the%20beginning%E2%80%94can%20make%20it%20easy%20for%20you%20to%20scale%20out%20your%20Postgres%20database%20at%20any%20time%2C%20to%20any%20scale.%20I%20like%20to%20call%20this%20being%20%E2%80%9Cscale-out-ready%E2%80%9D%20with%20Hyperscale%20(Citus)%20in%20Azure%20Database%20for%20PostgreSQL.%3C%2FSPAN%3E%3C%2FSPAN%3E%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CH2%20id%3D%22toc-hId-186344405%22%20id%3D%22toc-hId-186344405%22%3EWhat%20happens%20when%20you%20need%20to%20add%20worker%20nodes%20to%20your%20Hyperscale%20(Citus)%20Basic%20tier%3F%3C%2FH2%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EWhen%20you%E2%80%99re%20ready%20to%20scale%2C%20it%E2%80%99s%20easy%20to%20add%20worker%20nodes%20to%20your%20Hyperscale%20(Citus)%20Basic%20tier.%20After%20all%2C%20your%20data%20model%20is%20already%20distributed.%20And%20you%20don%E2%80%99t%20need%20to%20migrate%2C%20or%20to%20upgrade%2C%20or%20to%20go%20through%20any%20kind%20of%20hassle.%20You%20simply%20add%20more%20worker%20nodes%20in%20the%20portal%20and%20graduate%20from%20the%20Basic%20tier%20to%20the%20Standard%20tier%2C%20as%20we%E2%80%99ll%20do%20together%20below.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EWhen%20you%20add%20worker%20nodes%20to%20an%20Hyperscale%20(Citus)%20cluster%2C%20there%E2%80%99s%20no%20data%20on%20them%20yet.%20This%20is%20because%20all%20the%20shards%20are%20still%20on%20the%20old%20nodes%E2%80%94in%20this%20case%2C%20with%20Basic%20tier%2C%20all%20the%20data%20is%20on%20the%20single%20Hyperscale%20(Citus)%20node.%20So%2C%20initially%20all%20these%20new%20worker%20nodes%20will%20just%20be%20doing%20nothing.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EThis%20is%20where%20the%20Hyperscale%20(Citus)%20%3CA%20href%3D%22https%3A%2F%2Fwww.citusdata.com%2Fblog%2F2021%2F03%2F13%2Fscaling-out-postgres-with-citus-open-source-shard-rebalancer%2F%22%20target%3D%22_blank%22%20rel%3D%22noopener%20nofollow%20noreferrer%22%3Eshard%20rebalancing%20feature%3C%2FA%3E%20comes%20in.%20Shard%20rebalancing%20ensures%20that%20the%20shards%20are%20distributed%20fairly%20across%20all%20the%20nodes%20in%20the%20cluster.%20The%20Citus%20shard%20rebalancer%20does%20this%20by%20moving%20shards%20from%20one%20node%20to%20another.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EWhile%26nbsp%3Ba%20shard%20is%20being%20moved%2C%20all%20read%20and%20write%20queries%20can%20continue.%20In%20other%20words%2C%20Hyperscale%20(Citus)%20enables%20%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fhowto-hyperscale-scale-rebalance%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3Eonline%20rebalancing%3C%2FA%3E%E2%80%94or%20what%20some%20people%20call%20zero-downtime%20rebalancing%E2%80%94of%20the%20data.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EBelow%2C%20we%E2%80%99ll%20add%20new%20nodes%20to%20the%20Hyperscale%20(Citus)%20database%20cluster%20and%20rebalance%20shards%20across%20the%20cluster.%20We%E2%80%99ll%20just%20add%202%20worker%20nodes%2C%20but%20you%20can%20scale%20out%20your%20Hyperscale%20(Citus)%20server%20group%20to%20as%20many%20nodes%20as%20you%20think%20required.%3CBR%20%2F%3E%3CBR%20%2F%3E%3C%2FP%3E%0A%3CUL%3E%0A%3CLI%3E%3CSTRONG%3EGo%20to%20your%20Settings%20in%20the%20Azure%20portal%3C%2FSTRONG%3E%3A%20Look%20for%20the%20%3CSTRONG%3ECompute%20%2B%20storage%3C%2FSTRONG%3E%20tab%20in%20the%20Settings%20column%20on%20the%20left%20of%20the%20Azure%20portal%20screen.%3C%2FLI%3E%0A%3CLI%3E%3CSTRONG%3EChoose%20Standard%20tier%3C%2FSTRONG%3E%3A%20Choose%20Standard%20tier%3CSUP%3E%3CA%20href%3D%22%23ftn2%22%20target%3D%22_self%22%20rel%3D%22nofollow%20noopener%20noreferrer%22%20name%3D%22ftnref2%22%3E%5B2%5D%3C%2FA%3E%3C%2FSUP%3E%20to%20configure%20a%20multi-node%20cluster.%3C%2FLI%3E%0A%3CLI%3E%3CSTRONG%3EAdd%20worker%20nodes%3A%20%3C%2FSTRONG%3EUsing%20the%20slider%20in%20the%20screenshot%20below%2C%20pick%20any%20number%20of%20worker%20nodes%20you%20need.%20Remember%20that%20with%20Basic%20tier%2C%20you%20only%20have%20one%20Coordinator%20node%20and%20you%20don%E2%80%99t%20have%20any%20worker%20nodes.%20In%20this%20post%2C%20let%E2%80%99s%20add%202%20worker%20nodes%20below.%3C%2FLI%3E%0A%3CLI%3E%3CSTRONG%3EMake%20it%20so%3A%3C%2FSTRONG%3E%20Click%20the%20%E2%80%9CSave%E2%80%9D%20button%20at%20the%20bottom%20of%20the%20portal%20screen.%3C%2FLI%3E%0A%3C%2FUL%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3E%3CSPAN%20class%3D%22lia-inline-image-display-wrapper%20lia-image-align-inline%22%20image-alt%3D%22figure-5-screenshot-azure-portal-compute-storage-choosing-hyperscale-citus-standard-tier.png%22%20style%3D%22width%3A%20999px%3B%22%3E%3CIMG%20src%3D%22https%3A%2F%2Ftechcommunity.microsoft.com%2Ft5%2Fimage%2Fserverpage%2Fimage-id%2F273501i489873E3AB96D97D%2Fimage-size%2Flarge%3Fv%3Dv2%26amp%3Bpx%3D999%22%20role%3D%22button%22%20title%3D%22figure-5-screenshot-azure-portal-compute-storage-choosing-hyperscale-citus-standard-tier.png%22%20alt%3D%22Figure%205%3A%20Switch%20from%20Basic%20tier%20to%20Standard%20tier%20in%20Hyperscale%20(Citus)%20on%20the%20Azure%20Portal%E2%80%99s%20Compute%20%2B%20storage%20screen.%22%20%2F%3E%3CSPAN%20class%3D%22lia-inline-image-caption%22%20onclick%3D%22event.preventDefault()%3B%22%3EFigure%205%3A%20Switch%20from%20Basic%20tier%20to%20Standard%20tier%20in%20Hyperscale%20(Citus)%20on%20the%20Azure%20Portal%E2%80%99s%20Compute%20%2B%20storage%20screen.%3C%2FSPAN%3E%3C%2FSPAN%3E%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EOnce%20you%E2%80%99ve%20added%20the%20new%20Hyperscale%20(Citus)%20worker%20nodes%20and%20you%E2%80%99ve%20graduated%20your%20Basic%20tier%20to%20a%20Standard%20tier%2C%20you%20are%20ready%20to%20do%20some%20shard%20rebalancing.%20%3CBR%20%2F%3E%3CBR%20%2F%3E%3C%2FP%3E%0A%3CUL%3E%0A%3CLI%3E%3CSTRONG%3EYou%20decide%20when%20to%20initiate%20a%20shard%20rebalance%20operation%3C%2FSTRONG%3E%3A%20For%20maximum%20control%2C%20the%20choice%20of%20when%20to%20run%20the%20shard%20rebalancer%20is%20left%20to%20you.%20Hyperscale%20(Citus)%20does%20not%20automatically%20rebalance%20when%20you%20add%20new%20worker%20nodes.%3C%2FLI%3E%0A%3CLI%3E%3CSTRONG%3EWhat%20if%20you%20forget%20to%20rebalance%20shards%3C%2FSTRONG%3E%3F%20If%20you%20forget%20to%20rebalance%20you%20will%20have%20some%20empty%20worker%20nodes%20without%20any%20data.%20The%20good%20news%20is%2C%20you%20will%20see%20a%20recommendation%20to%20rebalance%20in%20the%20%E2%80%9CShard%20rebalancer%E2%80%9D%20tab%2C%20per%20the%20screenshot%20below.%3C%2FLI%3E%0A%3C%2FUL%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3E%3CSPAN%20class%3D%22lia-inline-image-display-wrapper%20lia-image-align-inline%22%20image-alt%3D%22figure-6-screenshot-azure-portal-shard-rebalancer-rebalancing-recommended.png%22%20style%3D%22width%3A%20999px%3B%22%3E%3CIMG%20src%3D%22https%3A%2F%2Ftechcommunity.microsoft.com%2Ft5%2Fimage%2Fserverpage%2Fimage-id%2F273502i3CDC393F9B13481E%2Fimage-size%2Flarge%3Fv%3Dv2%26amp%3Bpx%3D999%22%20role%3D%22button%22%20title%3D%22figure-6-screenshot-azure-portal-shard-rebalancer-rebalancing-recommended.png%22%20alt%3D%22Figure%206%3A%20Screenshot%20of%20the%20Azure%20portal%E2%80%99s%20Shard%20rebalancer%20screen%20when%20rebalancing%20is%20recommended.%20In%20this%20scenario%2C%20you%20can%20see%20that%20the%202%20newly-added%20worker%20nodes%20(blogtest-w0%20and%20blogtest-w1)%20each%20have%200%20MiB%20of%20data%E2%80%94they%20are%20empty!%22%20%2F%3E%3CSPAN%20class%3D%22lia-inline-image-caption%22%20onclick%3D%22event.preventDefault()%3B%22%3EFigure%206%3A%20Screenshot%20of%20the%20Azure%20portal%E2%80%99s%20Shard%20rebalancer%20screen%20when%20rebalancing%20is%20recommended.%20In%20this%20scenario%2C%20you%20can%20see%20that%20the%202%20newly-added%20worker%20nodes%20(blogtest-w0%20and%20blogtest-w1)%20each%20have%200%20MiB%20of%20data%E2%80%94they%20are%20empty!%3C%2FSPAN%3E%3C%2FSPAN%3E%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EYou%20can%20connect%20back%20to%20the%20coordinator%20via%3CCODE%20style%3D%22color%3A%20black%3B%22%3Epsql%3C%2FCODE%3Ewith%20the%20same%20connection%20string%20used%20earlier.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EAs%20I%20mentioned%20earlier%2C%20during%20the%20rebalance%20operation%2C%20all%20read%20and%20write%20queries%20can%20continue%20on%20the%20Hyperscale%20(Citus)%20database.%20In%20other%20words%2C%20your%20production%20workloads%20are%20not%20affected%20by%20the%20shard%20rebalancing.%20%3CBR%20%2F%3E%3CBR%20%2F%3E%3C%2FP%3E%0A%3CUL%3E%0A%3CLI%3E%3CSTRONG%3ERebalance%20shards%3C%2FSTRONG%3E%3A%20Once%20you%20are%20ready%2C%20rebalance%20the%20data%20with%20a%20single%20command%2C%3CCODE%20style%3D%22color%3A%20black%3B%22%3Erebalance_table_shards%3C%2FCODE%3E.%3C%2FLI%3E%0A%3C%2FUL%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CPRE%20class%3D%22lia-code-sample%20language-sql%22%3E%3CCODE%3E--%20move%20shards%20to%20new%20worker%20node(s)%0ASELECT%20rebalance_table_shards()%3B%0ANOTICE%3A%20%20Moving%20shard%20102008%20from%20c.blogtest.postgres.database.azure.com%3A5432%20to%20w1.blogtest.postgres.database.azure.com%3A5432%20...%0ANOTICE%3A%20%20Moving%20shard%20102009%20from%20c.blogtest.postgres.database.azure.com%3A5432%20to%20w0.blogtest.postgres.database.azure.com%3A5432%20...%0A....%0ANOTICE%3A%20%20Moving%20shard%20102028%20from%20c.blogtest.postgres.database.azure.com%3A5432%20to%20w1.blogtest.postgres.database.azure.com%3A5432%20...%3C%2FCODE%3E%3C%2FPRE%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EDuring%20the%20%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fhowto-hyperscale-scale-rebalance%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3Eshard%20rebalance%20operation%3C%2FA%3E%20in%20Hyperscale%20(Citus)%2C%20you%20can%20monitor%20its%20progress.%20To%20do%20that%2C%20follow%20the%20steps%20below%3A%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CUL%3E%0A%3CLI%3E%3CSTRONG%3EMonitor%20the%20shard%20rebalancer%3C%2FSTRONG%3E%3A%20Go%20to%20the%20Azure%20portal.%20Open%20the%20%3CSTRONG%3EShard%20rebalancer%3C%2FSTRONG%3E%20page%20in%20%3CSTRONG%3EServer%20group%20management%3C%2FSTRONG%3E.%20You%20will%20see%20the%20message%20that%20Rebalancing%20is%20underway.%20And%20you%20can%20follow%20the%20progress%20of%20the%20rebalance%20operation%20in%20this%20page.%3C%2FLI%3E%0A%3C%2FUL%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3E%3CSPAN%20class%3D%22lia-inline-image-display-wrapper%20lia-image-align-inline%22%20image-alt%3D%22figure-7-screenshot-azure-portal-shard-rebalancer-rebalancing-in-progress.png%22%20style%3D%22width%3A%20999px%3B%22%3E%3CIMG%20src%3D%22https%3A%2F%2Ftechcommunity.microsoft.com%2Ft5%2Fimage%2Fserverpage%2Fimage-id%2F273503i2166B7FD547D7FBC%2Fimage-size%2Flarge%3Fv%3Dv2%26amp%3Bpx%3D999%22%20role%3D%22button%22%20title%3D%22figure-7-screenshot-azure-portal-shard-rebalancer-rebalancing-in-progress.png%22%20alt%3D%22Figure%207%3A%20Follow%20the%20rebalance%20progress%20on%20the%20Azure%20Portal%E2%80%99s%20Shard%20Rebalancer%20screen%20while%20the%20rebalance%20is%20in%20progress.%22%20%2F%3E%3CSPAN%20class%3D%22lia-inline-image-caption%22%20onclick%3D%22event.preventDefault()%3B%22%3EFigure%207%3A%20Follow%20the%20rebalance%20progress%20on%20the%20Azure%20Portal%E2%80%99s%20Shard%20Rebalancer%20screen%20while%20the%20rebalance%20is%20in%20progress.%3C%2FSPAN%3E%3C%2FSPAN%3E%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3ENow%20that%20the%20shards%20have%20been%20distributed%20across%20the%20Hyperscale%20(Citus)%20cluster%2C%20your%20application%20can%20use%20the%20resources%20on%20the%20worker%20node(s)%20as%20well%20as%20the%20coordinator%20node.%20Other%20than%20the%20increased%20compute%2C%20memory%2C%20and%20disk%20available%20to%20your%20application%20though%E2%80%94from%20your%20application%E2%80%99s%20perspective%2C%20nothing%20has%20changed.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EAfter%20adding%202%20worker%20nodes%20to%20the%20Hyperscale%20(Citus)%20database%20cluster%2C%20and%20after%20rebalancing%20shards%20across%20the%20cluster%2C%20your%20application%20is%20still%20talking%20to%20the%20same%20Postgres%20database.%20Congratulations%2C%20you%20have%20seamlessly%20scaled%20out%20your%20Postgres%20database%20with%20Hyperscale%20(Citus)!%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CH2%20id%3D%22toc-hId--1621110058%22%20id%3D%22toc-hId--1621110058%22%3EWhy%20use%20Basic%20tier%20in%20Hyperscale%20(Citus)%20to%20shard%20on%20a%20single%20node%2C%20if%20you%20don't%20need%20to%20scale%20out%20Postgres%20right%20now%3F%3C%2FH2%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EYou%20may%20be%20wondering%3A%3C%2FP%3E%0A%3CBLOCKQUOTE%3E%0A%3CP%3E%E2%80%9CWhy%20use%20Hyperscale%20(Citus)%20if%20I%20don%E2%80%99t%20need%20to%20scale%20out%20Postgres%20right%20now%3F%E2%80%9D%3C%2FP%3E%0A%3C%2FBLOCKQUOTE%3E%0A%3CP%3EWell%2C%20if%20you%20think%20that%20your%20database%20is%20not%20going%20to%20grow%20in%20the%20future%20(and%20that%20your%20database%20will%20remain%20less%20than%20~100GB)%2C%20then%20%3CSTRONG%3Eone%20of%20the%20other%20deployment%20options%3C%2FSTRONG%3E%3CSTRONG%3E%20in%3C%2FSTRONG%3E%20our%20%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2F%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3EAzure%20Database%20for%20PostgreSQL%20managed%20service%3C%2FA%3E%E2%80%94such%20as%20Single%20Server%20or%20Flexible%20Server%E2%80%94is%20likely%20to%20handle%20your%20workload%20quite%20well.%20However%2C%20if%20you%20expect%20your%20application%20(and%20therefore%20your%20Postgres%20database)%20to%20grow%20over%20time%2C%20then%20the%20answer%20becomes%20relevant%20to%20you.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EIf%20you%20start%20small%20with%20Hyperscale%20(Citus)%E2%80%94by%20sharding%20Postgres%20on%20a%20single%20Hyperscale%20(Citus)%20node%20Basic%20tier%E2%80%94then%20as%20your%20application%20grows%20(in%20terms%20of%20users%20or%20activity%20or%20features%20or%20database%20size)%2C%20you%E2%80%99ll%20be%20able%20to%20easily%20add%20nodes%20and%20use%20the%20%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fhowto-hyperscale-scale-rebalance%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3Ezero%20downtime%20shard%20rebalancer%3C%2FA%3E%20on%20Hyperscale%20(Citus).%20Adding%20more%20nodes%20to%20your%20Hyperscale%20(Citus)%20server%20group%20will%20enable%20you%20to%20fit%20more%20data%20into%20memory%2C%20have%20higher%20I%2FO%20bandwidth%2C%20and%20provide%20more%20compute%20power%20(CPU)%20for%20your%20Postgres%20database.%20In%20other%20words%2C%20you%20will%20have%20all%20the%20means%20to%20keep%20your%20database%20performant%20for%20your%20application%20even%20as%20your%20application%20grows.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CH2%20id%3D%22toc-hId-866402775%22%20id%3D%22toc-hId-866402775%22%3EHow%20to%20prepare%20your%20Postgres%20database%20to%20be%20scale-out-ready%3F%3C%2FH2%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EThere%20are%20a%20few%20things%20you%20need%20to%20take%20into%20account%20to%20make%20Hyperscale%20(Citus)%20scale-out-ready%20with%20the%20new%20Basic%20tier.%20In%20scale-out%20architectures%2C%20the%20data%20is%20distributed%20based%20on%20a%20%3CA%20href%3D%22https%3A%2F%2Fwww.citusdata.com%2Fblog%2F2018%2F01%2F10%2Fsharding-in-plain-english%2F%22%20target%3D%22_blank%22%20rel%3D%22noopener%20nofollow%20noreferrer%22%3Esharding%20key%3C%2FA%3E%20(what%20we%20call%20a%20%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fconcepts-hyperscale-choose-distribution-column%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3Edistribution%20column%3C%2FA%3E%20in%20Citus.)%20The%20sharding%20key%20you%20choose%20will%20influence%20not%20only%20how%20the%20data%20gets%20distributed%2C%20but%20also%20what%20kind%20of%20query%20performance%20you%20will%20get.%20So%20a%20bit%20of%20up-front%20thinking%20about%20your%20query%20patterns%20and%20your%20data%20model%20can%20go%20a%20long%20way.%20For%20example%2C%20it%20is%20often%20not%20performant%20(or%20even%20possible)%20to%20enforce%20UNIQUE%20constraints%20on%20the%20Postgres%20columns%20that%20do%20not%20include%20distribution%20keys.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EIf%20you%20follow%20the%20data%20modeling%20best%20practices%20of%20a%20distributed%20database%20when%20using%20the%20Basic%20tier%20in%20Hyperscale%20(Citus)%20in%20Azure%20Database%20for%20PostgreSQL%2C%20you%20can%20easily%20scale%20out%20your%20database.%20The%20Citus%20best%20practices%20for%20scaling%20out%20your%20Postgres%20database%20seamlessly%20include%3A%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CUL%3E%0A%3CLI%3E%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fconcepts-hyperscale-choose-distribution-column%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3EPick%20the%20right%20distribution%20key%3C%2FA%3E%3C%2FLI%3E%0A%3CLI%3E%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fhowto-hyperscale-modify-distributed-tables%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3EUnderstand%20the%20different%20table%20types%3C%2FA%3E%20for%20Hyperscale%20(Citus)%2C%20including%20reference%20tables%20%26amp%3B%20distributed%20tables%3C%2FLI%3E%0A%3CLI%3E%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fhowto-hyperscale-modify-distributed-tables%23addingremoving-constraints%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3EAdjust%20primary%2Fforeign%20keys%20to%20include%20distribution%20key%3C%2FA%3E%3C%2FLI%3E%0A%3CLI%3E%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fconcepts-hyperscale-choose-distribution-column%23best-practices%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3EAdd%20distribution%20key%20filters%20and%20joins%20to%20the%20queries%3C%2FA%3E%20as%20much%20as%20possible%3C%2FLI%3E%0A%3CLI%3EBe%20aware%20of%20the%20few%20cross-shard%20%3CA%20href%3D%22https%3A%2F%2Fdocs.citusdata.com%2Fen%2Fstable%2Fdevelop%2Freference_workarounds.html%23sql-support-and-workarounds%22%20target%3D%22_blank%22%20rel%3D%22noopener%20nofollow%20noreferrer%22%3ESQL%20limitations%3C%2FA%3E%20which%20our%20Citus%2010%20open%20source%20team%20is%20working%20very%20hard%20to%20remove%20on%20each%20new%20%3CA%20href%3D%22https%3A%2F%2Fgithub.com%2Fcitusdata%2Fcitus%2Fblob%2Fmaster%2FCHANGELOG.md%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3ECitus%20release%3C%2FA%3E%3C%2FLI%3E%0A%3C%2FUL%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EIf%20you%20build%20your%20database%20conforming%20to%20the%20best%20practices%20outlined%20above%2C%20the%20promise%20of%20Hyperscale%20(Citus)%20is%20that%20you%E2%80%99ll%20be%20able%20to%20%3CA%20href%3D%22https%3A%2F%2Ftechcommunity.microsoft.com%2Ft5%2Fazure-database-for-postgresql%2Farchitecting-petabyte-scale-analytics-by-scaling-out-postgres-on%2Fba-p%2F969685%22%20target%3D%22_blank%22%3Escale%20out%20your%20Postgres%20database%3C%2FA%3E%20to%20some%20pretty%20large%20database%20cluster%20sizes.%20The%20key%20point%20here%20is%20that%20once%20you%20get%20used%20to%20the%20mindset%20of%20scale-out%20systems%2C%20you%E2%80%99ll%20realize%20how%20easy%20it%20is%20to%20follow%20the%20data%20modeling%20best%20practices.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3E%3CSPAN%20class%3D%22lia-inline-image-display-wrapper%20lia-image-align-inline%22%20image-alt%3D%22figure-8-diagram-citus-elicorn-single-node-then-scaling-out-transparent.png%22%20style%3D%22width%3A%20999px%3B%22%3E%3CIMG%20src%3D%22https%3A%2F%2Ftechcommunity.microsoft.com%2Ft5%2Fimage%2Fserverpage%2Fimage-id%2F273505iB132EBD7536978B8%2Fimage-size%2Flarge%3Fv%3Dv2%26amp%3Bpx%3D999%22%20role%3D%22button%22%20title%3D%22figure-8-diagram-citus-elicorn-single-node-then-scaling-out-transparent.png%22%20alt%3D%22Figure%208%3A%20Start%20small%20and%20cost-effectively%20with%20the%20new%20Hyperscale%20(Citus)%20%E2%80%9CBasic%20tier%E2%80%9D%20feature%20(in%20Preview)%2C%20and%20then%20if%20and%20when%20you%20need%20to%2C%20you%20can%20easily%20graduate%20to%20%E2%80%9CStandard%20tier%E2%80%9D%20by%20adding%20more%20Hyperscale%20(Citus)%20nodes.%22%20%2F%3E%3CSPAN%20class%3D%22lia-inline-image-caption%22%20onclick%3D%22event.preventDefault()%3B%22%3EFigure%208%3A%20Start%20small%20and%20cost-effectively%20with%20the%20new%20Hyperscale%20(Citus)%20%E2%80%9CBasic%20tier%E2%80%9D%20feature%20(in%20Preview)%2C%20and%20then%20if%20and%20when%20you%20need%20to%2C%20you%20can%20easily%20graduate%20to%20%E2%80%9CStandard%20tier%E2%80%9D%20by%20adding%20more%20Hyperscale%20(Citus)%20nodes.%3C%2FSPAN%3E%3C%2FSPAN%3E%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CH2%20id%3D%22toc-hId--941051688%22%20id%3D%22toc-hId--941051688%22%3EUsing%20Basic%20tier%20for%20your%20development%20%26amp%3B%20test%20environments%3C%2FH2%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EAnother%20interesting%20scenario%20for%20using%20Basic%20tier%20in%20Hyperscale%20(Citus)%20is%20with%20your%20development%20and%20test%20environments.%20As%20I%20mentioned%20earlier%2C%20the%20Basic%20tier%20does%20not%20have%20any%20limitations%20compared%20to%20Standard%20tier.%20So%2C%20for%20many%20users%2C%20Basic%20tier%20can%20be%20a%20practical%20and%20cost-effective%20way%20of%20provisioning%20development%20and%20testing%20databases.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CH2%20id%3D%22toc-hId-1546461145%22%20id%3D%22toc-hId-1546461145%22%3EAnd%20there%20are%20more%20Preview%20features%20available%20in%20Hyperscale%20(Citus)%3C%2FH2%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EAs%20I%20mentioned%20earlier%20in%20the%20blog%2C%20the%20Basic%20tier%20is%20currently%20in%20preview.%20In%20fact%2C%20there%20are%20several%20super%20useful%20%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fhyperscale-preview-features%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3EHyperscale%20(Citus)%20features%20in%20preview%3C%2FA%3E%2C%20including%3A%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CUL%3E%0A%3CLI%3E%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fconcepts-hyperscale-tiers%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3E%3CSTRONG%3EBasic%20tier%3C%2FSTRONG%3E%3C%2FA%3E%3CSTRONG%3E%3A%3C%2FSTRONG%3E%20Run%20a%20server%20group%20using%20only%20a%20coordinator%20node%20and%20no%20worker%20nodes.%20An%20economical%20way%20to%20do%20initial%20testing%20and%20development%20and%20handle%20small%20production%20workloads.%3C%2FLI%3E%0A%3CLI%3E%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fconcepts-hyperscale-versions%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3E%3CSTRONG%3EPostgreSQL%2012%20and%2013%3C%2FSTRONG%3E%3C%2FA%3E%3CSTRONG%3E%3A%3C%2FSTRONG%3E%20You%20can%20now%20choose%20among%20the%20latest%20Postgres%20versions%20for%20your%20Hyperscale%20(Citus)%20server%20group.%3C%2FLI%3E%0A%3CLI%3E%3CA%20href%3D%22https%3A%2F%2Fwww.citusdata.com%2Fblog%2F2021%2F03%2F05%2Fcitus-10-release-open-source-rebalancer-and-columnar-for-postgres%2F%22%20target%3D%22_blank%22%20rel%3D%22noopener%20nofollow%20noreferrer%22%3E%3CSTRONG%3ECitus%2010%3C%2FSTRONG%3E%3C%2FA%3E%3CSTRONG%3E%3A%20%3C%2FSTRONG%3EInstalled%20automatically%20on%20server%20groups%20running%20these%20Preview%20features%20and%20PostgreSQL%2013.%3C%2FLI%3E%0A%3CLI%3E%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fconcepts-hyperscale-columnar%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3E%3CSTRONG%3EColumnar%20storage%3C%2FSTRONG%3E%3C%2FA%3E%3CSTRONG%3E%3A%3C%2FSTRONG%3E%20Store%20selected%20tables'%20columns%20(rather%20than%20rows)%20contiguously%20on%20disk.%20Supports%20on-disk%20compression.%20Good%20for%20analytic%20and%20data%20warehousing%20workloads.%20(You%20might%20also%20find%20our%20open%20source%20blog%20%3CA%20href%3D%22https%3A%2F%2Fwww.citusdata.com%2Fblog%2F2021%2F03%2F06%2Fcitus-10-columnar-compression-for-postgres%2F%22%20target%3D%22_blank%22%20rel%3D%22noopener%20nofollow%20noreferrer%22%3Epost%20on%20Citus%2010%20columnar%20compression%3C%2FA%3E%20useful.)%3C%2FLI%3E%0A%3CLI%3E%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fhowto-hyperscale-read-replicas-portal%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3E%3CSTRONG%3ERead%20replicas%3C%2FSTRONG%3E%3C%2FA%3E%20(currently%20same-region%20only)%3A%20Any%20changes%20that%20happen%20to%20the%20primary%20server%20group%20get%20reflected%20in%20its%20replica%2C%20and%20queries%20against%20the%20replica%20cause%20no%20extra%20load%20on%20the%20original.%20Replicas%20are%20a%20useful%20tool%20to%20improve%20performance%20for%20read-only%20workloads.%3C%2FLI%3E%0A%3CLI%3E%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fconcepts-hyperscale-limits%23managed-pgbouncer-preview%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3E%3CSTRONG%3EManaged%20PgBouncer%3C%2FSTRONG%3E%3C%2FA%3E%3CSTRONG%3E%3A%3C%2FSTRONG%3E%20A%20connection%20pooler%20that%20allows%20many%20clients%20to%20connect%20to%20the%20server%20group%20at%20once%2C%20while%20limiting%20the%20number%20of%20active%20connections.%20Having%20an%20inbound%20PgBouncer%20satisfies%20connection%20requests%20while%20keeping%20the%20coordinator%20node%20running%20smoothly.%3C%2FLI%3E%0A%3C%2FUL%3E%0A%3CH2%20id%3D%22toc-hId--260993318%22%20id%3D%22toc-hId--260993318%22%3EThe%20Basic%20tier%20in%20Hyperscale%20(Citus)%20opens%20the%20door%20to%20new%20possibilities%3C%2FH2%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EWe%20are%20so%20excited%20to%20bring%20you%20a%20preview%20of%20the%20%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fconcepts-hyperscale-tiers%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3Enew%20Basic%20tier%20feature%3C%2FA%3E%20(in%20preview!)%20in%20Hyperscale%20(Citus)%20on%20Azure%20Database%20for%20PostgreSQL.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EAnd%20while%20%E2%80%9Copens%20the%20door%20to%20new%20possibilities%E2%80%9D%20may%20sound%20lofty%2C%20it%E2%80%99s%20true.%20Basic%20tier%20in%20Hyperscale%20(Citus)%20gives%20you%20a%20way%20to%20be%20%E2%80%9Cscale-out-ready%E2%80%9D%20on%20day%20zero%20in%20the%20cloud.%20If%20your%20application%20is%20already%20running%20on%20a%20single-node%20Postgres%2C%20you%20now%20have%20a%20way%20to%20adopt%20a%20sharded%20data%20model%20that%20allows%20you%20to%20scale%20your%20Postgres%20database%20as%20much%20as%20you%20need%20in%20the%20future.%20In%20essence%2C%20as%20your%20application%20grows%20and%20you%20need%20to%20scale%20out%2C%20you%20won%E2%80%99t%20be%20faced%20with%20any%20type%20of%20database%20migration%20challenges.%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3ESome%20useful%20links%20%26amp%3B%20resources%20when%20you%20are%20ready%20to%20try%20out%20Citus%3A%3C%2FP%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CUL%3E%0A%3CLI%3E%3CSTRONG%3EGetting%20started%3A%3C%2FSTRONG%3E%20This%20%3CA%20href%3D%22https%3A%2F%2Fwww.citusdata.com%2Fgetting-started%2F%22%20target%3D%22_blank%22%20rel%3D%22noopener%20nofollow%20noreferrer%22%3EGetting%20Started%20page%3C%2FA%3E%2C%20created%20by%20one%20of%20my%20teammates%2C%20has%20useful%20resources%20no%20matter%20which%20way%20you%20prefer%20to%20learn%3A%20reading%2C%20watching%2C%20or%20doing.%20It%E2%80%99s%20a%20%2B1%20for%20getting%20started%20with%20Citus%20open%20source%20as%20well%20as%20with%20Hyperscale%20(Citus)%20in%20the%20Azure%20Database%20for%20PostgreSQL%20managed%20service.%3C%2FLI%3E%0A%3CLI%3E%3CSTRONG%3EQuickstart%20documentation%20for%20Hyperscale%20(Citus)%3C%2FSTRONG%3E%3A%20These%20%3CA%20href%3D%22https%3A%2F%2Fdocs.microsoft.com%2Fazure%2Fpostgresql%2Fquickstart-create-hyperscale-basic-tier%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3EQuickstart%20docs%20for%20provisioning%20Basic%20tier%3C%2FA%3E%20in%20Hyperscale%20(Citus)%20are%20helpful.%20(And%20if%20you%20don%E2%80%99t%20yet%20have%20an%20Azure%20subscription%2C%20just%20create%20a%20%3CA%20href%3D%22https%3A%2F%2Fazure.microsoft.com%2Ffree%2F%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3Efree%20Azure%20account%3C%2FA%3E%20first.)%20%26nbsp%3B%3C%2FLI%3E%0A%3CLI%3E%3CSTRONG%3ECitus%20open%20source%3C%2FSTRONG%3E%3A%20Check%20out%20the%20%3CA%20href%3D%22https%3A%2F%2Fgithub.com%2Fcitusdata%2Fcitus%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3ECitus%20open%20source%20repo%3C%2FA%3E%20on%20GitHub%20to%20report%20issues%2C%20check%20out%20the%20source%20code%2C%20and%20learn%20more.%20To%20try%20Citus%20open%20source%2C%20you%20can%20follow%20the%20installation%20instructions%20on%20the%20GitHub%20%3CA%20href%3D%22https%3A%2F%2Fgithub.com%2Fcitusdata%2Fcitus%23getting-started%22%20target%3D%22_blank%22%20rel%3D%22noopener%20noreferrer%22%3EREADME%3C%2FA%3E%20or%20you%20can%20download%20Citus%20packages%20to%20try%20Citus%20locally.%20And%20you%20can%20find%20all%20the%20details%20about%20%3CA%20href%3D%22https%3A%2F%2Fwww.citusdata.com%2Fblog%2F2021%2F03%2F20%2Fsharding-postgres-on-a-single-citus-node%2F%22%20target%3D%22_blank%22%20rel%3D%22noopener%20nofollow%20noreferrer%22%3Esharding%20Postgres%20on%20a%20single%20Citus%20node%3C%2FA%3E%20in%20my%20open%20source%20blog%20post%2C%20too.%3C%2FLI%3E%0A%3CLI%3E%3CSTRONG%3EDeep%20dive%20into%20Citus%20internals%3C%2FSTRONG%3E%3A%20If%20you%20want%20to%20understand%20the%20internals%20of%20how%20Citus%20shards%20Postgres%20as%20an%20extension%2C%20%3CA%20href%3D%22https%3A%2F%2Fwww.citusdata.com%2Fblog%2F2021%2F04%2F10%2Ftalk-at-cmu-how-citus-distributes-postgresql-via-extension-apis%2F%22%20target%3D%22_blank%22%20rel%3D%22noopener%20nofollow%20noreferrer%22%3Ethis%20recent%20blog%20post%3C%2FA%3E%20about%20one%20of%20my%20teammate%E2%80%99s%20Citus%20talks%20at%20Carnegie%20Mellon%20(with%20%3CA%20href%3D%22https%3A%2F%2Fyoutu.be%2FX-aAgXJZRqM%22%20target%3D%22_blank%22%20rel%3D%22noopener%20nofollow%20noreferrer%22%3Evideo%20recording%20by%20the%20CMU%20team%3C%2FA%3E)%20is%20a%20great%20place%20to%20start.%3C%2FLI%3E%0A%3C%2FUL%3E%0A%3CP%3E%26nbsp%3B%3C%2FP%3E%0A%3CP%3EAnd%20if%20you%20have%20any%20questions%20about%20our%20Hyperscale%20(Citus)%20option%20in%20our%20Azure%20Database%20for%20PostgreSQL%20managed%20service%2C%20my%20product%20teammates%20would%20love%20to%20hear%20from%20you.%20You%20can%20always%20reach%20out%20to%20us%20over%20email%20at%20%3CA%20href%3D%22mailto%3AAskAzureDBforPostgreSQL%40service.microsoft.com%3Fsubject%3DBasic%2520tier%2520for%2520Hyperscale%2520(Citus)%2520Preview%2520blog%2520post%22%20target%3D%22_blank%22%20rel%3D%22noopener%20nofollow%20noreferrer%22%3EAskAzureDBforPostgreSQL%3C%2FA%3E.%20Or%20follow%20us%20on%20Twitter%20%3CA%20href%3D%22https%3A%2F%2Ftwitter.com%2FAzureDBPostgres%22%20target%3D%22_blank%22%20rel%3D%22noopener%20nofollow%20noreferrer%22%3E%40AzureDBPostgres%3C%2FA%3E.%20We%E2%80%99d%20love%20to%20know%20what%20you%20think.%3C%2FP%3E%0A%3CHR%20%2F%3E%0A%3CSECTION%20role%3D%22doc-endnotes%22%3E%0A%3CP%3E%3CSTRONG%3EFootnotes%3C%2FSTRONG%3E%3C%2FP%3E%0A%3COL%3E%0A%3CLI%20id%3D%22ftn1%22%20role%3D%22doc-endnote%22%3EIn%20the%20East%20US%20region%20on%20Azure%2C%20the%20cost%20of%20a%20Hyperscale%20(Citus)%20Basic%20tier%20with%202%20vCores%2C%208%20GiB%20total%20memory%2C%20and%20128%20GiB%20of%20storage%20on%20the%20coordinator%20node%20is%20%240.27%2Fhour%20or%20~%24200%2Fmonth.%20%3CA%20role%3D%22doc-backlink%22%20href%3D%22%23ftnref1%22%20target%3D%22_self%22%20rel%3D%22nofollow%20noopener%20noreferrer%22%3E%E2%86%A9%3C%2FA%3E%3CBR%20%2F%3E%3CBR%20%2F%3E%3C%2FLI%3E%0A%3CLI%20id%3D%22ftn2%22%20role%3D%22doc-endnote%22%3EThe%20minimum%20Hyperscale%20(Citus)%20coordinator%20size%20for%20Standard%20tier%20is%204%20vCores%20with%2016%20GiB%20memory%20and%20512%20GiB%20disk.%20So%20if%20you%20have%20the%20smallest%20Basic%20tier%20configuration%2C%20then%20switching%20to%20Standard%20tier%20will%20increase%20your%20Coordinator%20node%20size%20from%202%20vCores%20to%204vCores%20(or%20more%20if%20you%20choose)%20and%20from%20128%20GiB%20storage%20to%20512%20GiB%20storage%20(or%20more%20if%20you%20choose.)%20If%20your%20coordinator%20node%20needs%20to%20change%20size%20when%20moving%20from%20Basic%20to%20Standard%20tiers%2C%20your%20coordinator%20will%20need%20to%20be%20restarted.%20Restarting%20the%20coordinator%20requires%20a%20brief%20downtime.%20You%E2%80%99ll%20see%20a%20warning%20before%20the%20restart%20happens%2C%20so%20you%20can%20confirm%20that%20the%20downtime%20is%20OK%20at%20that%20moment.%20You%20would%20avoid%20the%20restart%20if%20your%20Basic%20Tier%20already%20had%204%20vCores%20(or%208%20vCores)%20and%20512%20GiB%20storage.%20%3CA%20role%3D%22doc-backlink%22%20href%3D%22%23ftnref2%22%20target%3D%22_self%22%20rel%3D%22nofollow%20noopener%20noreferrer%22%3E%E2%86%A9%3C%2FA%3E%3C%2FLI%3E%0A%3C%2FOL%3E%0A%3C%2FSECTION%3E%3C%2FLINGO-BODY%3E%3CLINGO-TEASER%20id%3D%22lingo-teaser-2275672%22%20slang%3D%22en-US%22%3E%3CP%3E%3CSPAN%20class%3D%22lia-inline-image-display-wrapper%20lia-image-align-inline%22%20image-alt%3D%22single-blue-postgres-elephant-surrounded-by-lighter-elephants-1200x630.jpg%22%20style%3D%22width%3A%20999px%3B%22%3E%3CIMG%20src%3D%22https%3A%2F%2Ftechcommunity.microsoft.com%2Ft5%2Fimage%2Fserverpage%2Fimage-id%2F273452i67A2CC7142DC3233%2Fimage-size%2Flarge%3Fv%3Dv2%26amp%3Bpx%3D999%22%20role%3D%22button%22%20title%3D%22single-blue-postgres-elephant-surrounded-by-lighter-elephants-1200x630.jpg%22%20alt%3D%22single-blue-postgres-elephant-surrounded-by-lighter-elephants-1200x630.jpg%22%20%2F%3E%3C%2FSPAN%3E%3C%2FP%3E%0A%3CP%3EBasic%20tier%20makes%20it%20easy%20to%20get%20started%20with%20Hyperscale%20(Citus)%20%26amp%3B%20easy%20to%20adopt%20a%20distributed%20data%20model%20from%20the%20beginning.%3C%2FP%3E%3C%2FLINGO-TEASER%3E%3CLINGO-LABS%20id%3D%22lingo-labs-2275672%22%20slang%3D%22en-US%22%3E%3CLINGO-LABEL%3EAzure%20Database%20for%20PostgreSQL%3C%2FLINGO-LABEL%3E%3CLINGO-LABEL%3ECitus%3C%2FLINGO-LABEL%3E%3CLINGO-LABEL%3EPostgres%20extensions%3C%2FLINGO-LABEL%3E%3CLINGO-LABEL%3EPostgreSQL%3C%2FLINGO-LABEL%3E%3CLINGO-LABEL%3Escalability%3C%2FLINGO-LABEL%3E%3C%2FLINGO-LABS%3E
Co-Authors
Version history
Last update:
‎Apr 17 2021 11:40 AM
Updated by: