Google I/O: AlloyDB combines PostgreSQL with GCP

Choice is always a good thing, and starting today, businesses facing data overflow issues – and that affects just about every business – will have a new cloud-based option for storing their files, data, metadata and logs.

This choice is offered by Google Cloud Platform, which unveiled today at the Google I/O 2022 virtual conference a new parallel database, AlloyDB, built on the popular open-source PostgreSQL database, which is a standard choice for developers for over three decades. The new database could therefore be both familiar and brand new to many of its users.

According to Gartner Research, 75% of all global databases are expected to be running in the cloud this year. Google AlloyDB should be able to support much of the data load created by the explosion of the e-commerce and social media sectors, as well as all the new applications and data that will fill digital vaults in preparation for the environment. Web3.

What AlloyDB brings to the market

Two of the main differentiators that AlloyDB brings are speed and more predictable pricing, Andi Gutmans, Google’s vice president of databases, told ZDNet.

“We can actually run analytic queries up to 100 times faster than the free software Postgres,” Gutmans said. “On the transactional side, based on our benchmarks, we’re about four times faster than open-source Postgres, and about twice as fast as Amazon’s equivalent offering (Aurora), so we’re working hard to ensure performance optimal on both the transactional and analytical side. This means that customers who want to do real-time fraud detection, real-time recommendations, real-time inventory management – they can actually do a lot of of this directly in their operational store.”

As for pricing, storing data in the cloud has always been an inexact science. Most cloud computing providers, such as Microsoft Azure, AWS, Dell, Oracle, and others, charge users for data egress, compute, IOPS, and other services.

So Google decided to take the risk itself and simplify the pricing model, Gutmans said. No access or IOPS fees will be charged for AlloyDB, he said.

“From the beginning, customers pay for compute and for the storage they use, they pay for IOPS usage,” Gutmans said. “That was actually one of the biggest pain points we heard from some customers – that it was about 60% of their bill. When it comes to IOPS, they didn’t feel like It’s really hard to manage because customers can’t really predict the cost, it depends on how much data is in memory, how much data is in storage, etc. So we dropped that charge.

“Our goal was really to make this computing experience a breeze – like an autopilot.”

Expected use cases of AlloyDB

AlloyDB is aimed at DBAs with code stacks that use a full database offering options such as ACID-compliant transactions and stored procedures (atomicity, consistency, isolation, and durability). Gutmans told ZDNet he expects AlloyDB to quickly enter the market and directly compete with conventional databases from Oracle, IBM or Microsoft by offering the necessary functionality in a package. cloud native which will be easier to control. CFOs will also appreciate this part.

Here’s how AlloyDB differentiates itself from its competitors, according to Mr. Gutmans:

  • Unlike Oracle, AlloyDB supports PostgreSQL. Additionally, AlloyDB supports both Postgres implementation and automation features such as autoscaling. It delivers four times faster performance than standard PostgreSQL for transactional workloads, and achieves a high availability SLA of up to 99.99%, according to Gutmans.
  • AlloyDB delivers more than twice as fast processing than Amazon Aurora, Gutmans said. Also, AlloyDB has a more user-friendly pricing model as shown above. Unlike Amazon Aurora, AlloyDB does not charge for I/O, which can be a significant source of unpredictable and hard-to-control costs – up to 60% of your total bill for transactional workloads.
  • AlloyDB is a new option for enterprises looking to move away from proprietary databases, as well as PostgreSQL users with demanding, high-end applications. It enables users to cost-effectively modernize their proprietary databases and scale their mission-critical workloads, Gutmans said.

In technical terms, says Gutmans, AlloyDB is:

  • an intelligent, database-optimized storage service;
  • an optimized database engine that is 100% compatible with PostgreSQL; and
  • a service with built-in autopilot capabilities, including built-in integration with Vertex AI, which allows users to invoke models directly within a query or transaction.

AlloyDB also has a built-in columnar accelerator that can run analytical queries faster than the standard version of PostgreSQL, Gutmans said. It is a flexible tool for developers thanks to the combination of PostgreSQL and the open infrastructure of Google Cloud. With this, developers will be able to use AlloyDB to quickly build applications and scale them, all with their existing open-source skills, Gutmans said.

All about Google I/O 2022

Source: “ZDNet.com”

Leave a Comment