Cloud Hosting

Google is enabling AlloyDB, the PostgreSQL-compatible database, to run anywhere, including public cloud, on-premises servers, edge computing environments and even developer laptops. Branded as AlloyDB Omni, the engine is the same as AlloyDB, the cloud-based managed database announced last year.

Elephant

pixabay

AlloyDB Omni promises compatibility with PostgreSQL, enhanced performance and support delivered by Google Cloud. Compared to the standard, open source PostgreSQL, AlloyDB Omni delivers 2x faster performance and 100x faster analytical queries. This is possible due to how Google has tuned, enhanced and optimized the database engine.

By analyzing a query's components, such as subqueries, joins, and filters, the AlloyDB Omni index advisor reduces the guesswork involved in tuning query performance. By periodically analyzing the workload on the database, it finds queries that could benefit from indexes, and suggests new indexes that could significantly improve query performance.

The other unique feature of AlloyDB Omni includes a columnar engine, which keeps frequently accessed data in an in-memory columnar format for quicker scans, joins, and aggregations. AlloyDB Omni automatically arranges the data and selects between columnar and row-based execution plans using machine learning. This capability delivers better performance without recreating queries targeting different formats.

AlloyDB Omni is packaged as a set of containers that can be deployed in a Debian-based or a Red Hat Enterprise Linux host. In its technical preview, Google is providing a set of shell scripts to automate the deployment. However, there is no guidance on deploying AlloyDB Omni in a Kubernetes cluster through Helm Chart or an operator. This may change when the software moves towards the general availability.

Google recommends deploying AlloyDB Omni on a machine or a VM with at least two CPUs and 16GB of memory. The machine should have Docker and Google Cloud SDK installed to pull the images of AlloyDB from Google Cloud Container Registry and the shell scripts uploaded to Google Cloud Storage. On a machine with prerequisites installed, it takes a couple of minutes to get AlloyDB Omni up and running.

Interestingly, Google doesnt mention Anthos as the preferred infrastructure for deploying AlloyDB Omni. Though the software is packaged as containers, it can run on any Linux machine with Docker installed.

AlloyDB Omni also supports the creation of read replicas - dedicated database servers optimized for read-only access. A replica server provides a read-only clone of the primary database server while continuously updating its own data to reflect changes made to the primary server's data. Read replicas significantly increases the throughput and availability of the database.

Google is investing in AlloyDB Omni to attract customers migrating their databases from legacy versions of Oracle and Microsoft SQL Server. With 100% compatibility with PostgreSQL, customers can take advantage of the migration tools and the expertise available in the ecosystem. The other use case is running an optimized database at the edge. Customers can ingest IoT device data into AlloyDB for querying and analyzing the telemetry data of various sensors. Similar to BigQuery Omni, enterprises can run a Google Cloud-managed database in other cloud environments such as AWS and Azure. It will simplify the integration of data services while reducing the bandwidth cost involved in moving the data across clouds.

Google is not the only public cloud provider to bring a cloud-based managed database to multicloud and on-premises. Microsoft announced Azure Arc-enabled SQL Server and Azure Arc-enabled PostgreSQL in 2020. Based on Azure Arc, Microsoft has packaged these databases as Kubernetes deployments. Enterprises with Arc-enabled Kubernetes can easily deploy SQL Server and PostgreSQL on their clusters.

Scaling AlloyDB Omni to the cloud-based version is straightforward. Like any other migration, customers can export the data in a CSV, DMP or SQL format and import that data into an AlloyDB instance running in Google Cloud. For lift-and-shift scenarios, Google recommends using the Database Migration Service, which is currently in preview.

With a clear migration plan to the cloud-based AlloyDB based on the recently announced Database Migration Service, Google hopes to drive the adoption of its Data Cloud through AlloyDB Omni.

Janakiram MSV is an analyst, advisor and an architect at Janakiram & Associates. He was the founder and CTO of Get Cloud Ready Consulting, a niche cloud migration and cloud operations firm that got acquired by Aditi Technologies. Through his speaking, writing and analysis, he helps businesses take advantage of the emerging technologies.

Janakiram is one of the first few Microsoft Certified Azure Professionals in India. He is one of the few professionals with Amazon Certified Solution Architect, Amazon Certified Developer and Amazon Certified SysOps Administrator credentials. Janakiram is a Google Certified Professional Cloud Architect. He is recognised by Google as the Google Developer Expert (GDE) for his subject matter expertise in cloud and IoT technologies. He is awarded the title of Most Valuable Professional and Regional Director by Microsoft Corporation. Janakiram is an Intel Software Innovator, an award given by Intel for community contributions in AI and IoT. Janakiram is a guest faculty at the International Institute of Information Technology (IIIT-H) where he teaches Big Data, Cloud Computing, Containers, and DevOps to the students enrolled for the Master's course. He is an Ambassador for The Cloud Native Computing Foundation.

Janakiram was a senior analyst with Gigaom Research analyst network where he analyzed the cloud services landscape. During his 18 years of corporate career, Janakiram worked at world-class product companies including Microsoft Corporation, Amazon Web Services and Alcatel-Lucent. His last role was with AWS as the technology evangelist where he joined them as the first employee in India. Prior to that, Janakiram spent over 10 years at Microsoft Corporation where he was involved in selling, marketing and evangelizing the Microsoft application platform and tools. At the time of leaving Microsoft, he was the cloud architect focused on Azure.

See more here:
Google Brings PostgreSQL-Compatible AlloyDB To Multicloud, Data Centers And The Edge - Forbes