PRIMARY: XtremIO, VMAX and ScaleIO
Considering the key requirements for scalability, flexibility, low latency and integration into orchestration/automation solutions, the tightly coupled scale-out architecture of XtremIO is the correct architecture choice for this workload.
Key production workload requirements include scale-out IOPS with consistent ultra-low latency regardless of workload mix or dataset size.
The ability to maintain multiple writable snapshot instances with little to no impact on the array, or a desire to lower the total cost of ownership through inline data reduction, compression and space-efficient writable snapshots across multiple instances makes the XtremIO all-flash array is the platform of choice. With the ability to scale out to 6 X-Bricks, delivering 120TB of capacity (before considering the impact of data reduction and copy services efficiencies) and proven 99.999+% availability, XtremIO can scale to fit all but the largest DBaaS deployments.
If the key production workload requirements are focussed on reliability, availability, and serviceability (RAS), specifically requiring zero RTO, or deep integration into orchestration and automation solutions specific to a particular DB offering, the platform of choice for these workloads is VMAX. With a bulletproof reputation and high levels of reliability availability, and supportability (RAS), VMAX offers zero RPO using SRDF synchronous DR replication. Leveraging SSD Tiers and large global memory, VMAX is able to deliver consistent high performance for these critical workloads.
Not all database workloads are created equal. As such, they don’t need Flash (AFA) technology in every instance. EMC ScaleIO, a software-defined block storage platform, augments database workloads by supporting multiple storage media (such as PCIe flash cards, SSDs, and HDDs) offering enterprise customers flexibility with an eye on cost containment while delivering SLA-based services. From a service provider perspective, ScaleIO can support critical functions like traditional RDBMS and NoSQL workloads, volume mobility, handle rebuilds faster and consistent point-in-time snapshots. Other revenue generating ScaleIO functionality includes agility in deployment, partitioning, thin provisioning, IOPS limiter, multi-tenancy, and QoS controls. ScaleIO is an ideal low ‘cost to serve’ option while delivering an innovative software-defined storage (SDS) platform that supports both enterprise (Oracle, SAP, Microsoft) and modern (next gen workloads like NoSQL – MongoDB, Cassandra, Couchbase as well as platforms like Openstack and Docker) applications.
The Spiderchart below shows the design characteristics of the platform(s) in relation to the workload requirements. Note overall alignment and relative strengths related to the target workload.