MemSQL extends our operational data platform with an on-demand, elastic cloud service, and new features to support Tier 1 workloads.
MemSQL is proud to announce two exciting new product releases today: MemSQL Helios, our on-demand, elastic cloud database-as-a-service, and MemSQL 7.0 Beta 2, the next major release of our database engine, featuring MemSQL SingleStore™ – a breakthrough new way of managing data – and new features to fully support Tier 1, system of record workloads.
With MemSQL Helios, you get instant, effortless access to the world’s fastest, most scalable data platform for operational analytics, machine learning, and AI, with platform operations handled by MemSQL. Helios is a fully managed cloud service that provides you with instant access to our best-in-class operational database — on demand and at elastic scale — in public cloud environments around the world.
MemSQL 7.0, which will be generally available in the coming months, brings two exciting new advance advances: MemSQL SingleStore and key features to make MemSQL “system of record” capable for Tier 1 workloads.
Data platform spending is seeing explosive growth as new applications are built to support real-time decisions, predictive analytics, and automation leveraging machine learning and AI. With this growth, and the desire for greater flexibility and cost control, more and more data workloads are moving to the cloud; see Gartner’s perspective in this recent analyst research.
The vast majority of MemSQL’s customers are already deploying our database in the cloud, typically using deployment automation or container infrastructure. Many have chosen MemSQL as the cloud migration path from older, on-premises operational database implementations.
The availability of MemSQL Helios, along with the imminent release of SingleStore and the new system of record capabilities, make MemSQL the ideal choice for both new applications and to move legacy operational databases to the cloud.
Announcing MemSQL Helios
MemSQL Helios has the best-in-class speed, scale, and robust features you have come to expect from MemSQL, but now available on demand. Customers can focus on using their data to build breakthrough applications and analytical systems instead of mundane infrastructure management and tuning.
With MemSQL Helios, there’s no software to deploy or manage. The most important steps in getting started are choosing an initial deployment size from a pull-down menu, then clicking a button to initiate deployment. Compute and storage resources are automatically assigned, the necessary software is installed and configured, and clusters are created, ready to store data — all in a few minutes. High availability is built in, with MemSQL handling data backup and, if needed, restore operations.
Benefits of MemSQL Helios include:
- Effortless deployment and management. As we have all come to expect from cloud services, deployment and upgrades are built in. With MemSQL Helios you get the full benefits and capabilities of the MemSQL data platform without having to worry about deployment, management, or maintenance. There’s no need to rack servers, script deployments, or manage VMs.
- Avoid cloud lock-in through multi-cloud flexibility. Helios is available today on Amazon Web Services and Google Cloud Platform. MemSQL operates exactly the same whether deployed on-premises on bare metal, across on-premises or cloud infrastructure, using the MemSQL Kubernetes Operator, or within the Helios service. You can use MemSQL to support a broad set of operational and analytical use cases, allowing for a simple, single platform across applications, analytical systems, and cloud deployments.
- Superior TCO. Compared to both legacy databases and proprietary databases from the cloud service providers, MemSQL Helios offers superior total cost of ownership (TCO). MemSQL offers high performance, scalability, ANSI SQL support, and the ability to replace traditional databases like Oracle Exadata and SAP Hana, at a fraction of the cost. When compared to the proprietary databases offered on Amazon Web Services and Google Cloud Platform, MemSQL’s unique architecture and high-performance query engine mean that many operational analytics workloads run with far less resource consumption, offering significant cost savings.
MemSQL Helios is available in limited public preview today. The service is secure, stable, and ready for your production workloads. There is a time-limited trial open to all users; however, Helios is initially available only to a limited number of customers to purchase. Try it for yourself now, or request a technical deep dive with a product specialist. If you are interested in purchasing MemSQL Helios immediately, please contact us to request an invitation.
Also, see our MemSQL Intro playlist on YouTube.
Introducing SingleStore: Breakthrough Data Management from MemSQL
With legacy databases, customers had their data in silos: transaction data in rowstore tables, analytics and data warehousing systems using columnstore tables, and extract, transform, and load (ETL) operations to bridge the gap. MemSQL brings rowstore and columnstore tables together in a single database; ETL is eliminated, and SQL queries can combine data from both types of tables. With SingleStore, in future versions of MemSQL, we will eliminate the need to choose one table type or another. The system will optimize storage and data access for you.
MemSQL is seeking to eliminate data duplication, reduce complexity, and cut total cost of ownership with the launch of MemSQL SingleStore, a new breakthrough in data architecture. In the future, SingleStore will offer the fastest possible performance, at the lowest possible cost, for every kind of workload – transactional, analytical, and hybrid – by storing all data in a single table type.
In MemSQL 7.0, SingleStore is delivered through improvements to both rowstore and columnstore tables, allowing each to handle workloads that previously only worked well on the other. Rowstore tables – used for transactions, for seeking a few rows of data, and for analytics on rapidly-changing data – get null elimination via the new sparse compression feature. In null elimination, fields that are subject to sparse compression are labeled as SPARSE when the table is created.
For each sparse field, a flag is created, half a byte wide. When the field’s value is null, the flag is set by the database software, and no value is stored for that column, saving that amount of storage – often 4, 8, or 16 bytes per null. Null elimination reduces the memory footprint of many rowstore tables by 50% or more, cutting memory usage and costs by half or more.
Also in MemSQL 7.0, columnstore tables – highly useful for most analytics purposes, but much harder to update efficiently than rowstore tables – get multiple secondary indexes, allowing fast seeks on multiple access paths, and locking at row level, which enables higher concurrency of updates and deletes.
With the initial implementation of MemSQL SingleStore delivered in MemSQL 7.0, you can use columnstore for more workloads than you could before – workloads that previously required rowstore. Using columnstore for these additional workloads means lower total cost of ownership (TCO), while still meeting service level agreements (SLAs) and user expectations. In addition, for the workloads that still require the updating capabilities and even faster performance of rowstore, compression lowers TCO for those workloads as well.
New Features Enable “System of Record” Capability
MemSQL 7.0 also introduces two key availability features which combine to enable Tier 1, system of record workloads: much faster sync replication, turned on by default, and incremental backup.
Fast sync replication introduces a high-availability replication option — the ability to always keep a replicated, live copy of your database as it’s receiving updates. Together with the availability of sync durability (copy to disk required for an update to be logged), long supported in MemSQL, sync replication enables the level of data safety required for Tier 1, system of record workloads.
Incremental backup extends the software’s capabilities beyond the ability to do full backups, to include regular backups of changed data from select time frames, and the ability to restore reliably from the previous full backup plus incremental backups.
Together, the addition of the ability to regularly backup recently changed data, and the ability to synchronously replicate your data, mean that MemSQL can now be trusted for critical data workloads.
Get Started Today
MemSQL Helios is secure, stable, and ready for production immediately. The service is priced based on consumption, with options for both on-demand purchase and discounted prepaid subscriptions. Not all deployment regions are supported initially.
Existing MemSQL customers can download MemSQL 7.0 Beta 2 to try out our new system of record features and SingleStore. If you are interested in evaluating MemSQL 7.0, please contact us to request access. MemSQL 7.0 will be generally available in the cloud and for download later this year. Sign up for our upcoming Helios webinar to learn more.
Also, see our MemSQL Intro playlist on YouTube.