Percona/Openstack Live 2015: Solve for Scale
Last week found the SolidFire team in Santa Clara, CA at the Percona Live conference, the world’s largest dedicated MySQL conference run by Percona, the premier MySQL professional services organization. This year Percona added a parallel event, OpenStack Live, bringing together database and DevOps professionals in an expanded ecosystem environment to share knowledge and best practices and explore the next iteration of open source platform architecture. Database as a Service (DBaaS) was a prominent topic, and the various crossover sessions were very well attended. The common theme of solving for scale and rapid growth in MySQL environments wound its way through many of the conference sessions.
One observation I made was that the MySQL community showed a lot of interest in platform technology — storage technologies in particular. Many attendees were database focused, but were also responsible for other areas of their platforms, owning the role of DB/DevOps experts. The topics of database consolidation, platform automation, and software-defined infrastructure resonated well. Technical experts and platform owners were keenly interested in innovative ways to address the problems of scale, management, and efficiency in the context of MySQL.
One of the strongest keynote and session themes was DBaaS. From VMware’s vCloudAir DBaaS initiative to Tesora’s work with Trove in OpenStack, the focus was on how to scale databases in an automated way, in a world where data volume and the requirements for data management have vastly outpaced the human element of hands-on database and system administration. The ability to integrate SolidFire directly with both VMware and OpenStack was pivotal in many of the discussions that I was pleased to have with technologists and business executives alike.
I enjoyed delivering a session on “Scaling MySQL Through Consolidation,” walking the audience through various considerations of managing MySQL at scale, as well as demonstrating examples of how database operations can evolve by leveraging a modern storage layer, such as SolidFire. I demonstrated this category of scaling techniques during the Lighting Talk round at the MySQL Community Event on April 15th.
As compute platforms get more flexible, both in public and private clouds, it becomes increasingly possible to turn the traditional model of database platform management on its head. Rather than seeing the data layer as the child or property of the compute layer, you can now make the compute layer an attribute of the dataset.
With data gravity and ‘stickiness’ being the operational bottleneck, you can pivot on your storage layer and use the capabilities of modern shared storage systems, combined with virtualization, to flip that model. I refer to this as the Storage Jiu Jitsu technique. Requisite storage capabilities include instant snapshots and quick volume cloning, as well as data deduplication for very efficient use of high-performance shared-storage platforms.
Quality of Service allocation is the key to segregating performance resources between storage volumes, allowing you to put production MySQL database systems on the same storage array as dev, test, and other secondary systems without fear of resource conflict (commonly known as the “noisy neighbor” effect). Scale-out architectures such as SolidFire are also a great development in storage, allowing for seamless growth without the need for huge migrations and massive capital poured into storage system refreshes.
All in all, this was a tremendously valuable conference. The highly technical audience comprised high-level influencers and decision makers to whom performance and scale were exceedingly relevant issues, and who were well versed in the relationship between platform, database performance, and storage.
What will happen in the MySQL and OpenStack communities between now and Percona/Openstack Live 2016? Personally, I’m eagerly looking forward to being part of the continuing evolution of open source database platforms and accepting the challenge: Solve for Scale.