Digital transformation has created a demand for real-time data that is driving the expansion of real-time analytics and AI/ML applications. This demand has exposed a performance gap between what current solutions can deliver versus what they need to be able to deliver. Real-time applications rely on the available in-memory space to properly perform and yesterday’s technology just doesn’t have the capacity, performance and availability that today’s applications require.
MemVerge Memory Machine virtualizes persistent memory (PMEM) such as Intel’s Optane DC PMEM by allowing applications to access it as both volatile memory and persistent memory without rewriting applications. In addition, Memory Machine provides data services such as in-memory snapshots and replication, addressing application fault tolerance and data durability challenges when applications are deployed on persistent memories.
The target audience for this lab are customers who are looking for technology that is scalable and highly available for their mission-critical real-time workloads that require an in-memory solution.
The solution we are using to demonstrate the MemVerge Memory Machine is comprised of two Dell PowerEdge R740 servers with:
- Intel Xeon Gold 6240 processors
- 12 x 32Gb DDR-4 Memory Modules
- 12 x 256Gb Intel Optane PMem
- 2 x 10GBe Intel X710 NICs
- 6 x 960Gb SAS SSDs
The software currently installed for this solution includes:
- CentOS 8 for the operating system
- MemVerge Memory Machine V1.0
- Redis 5.0.7
- memtier_benchmark 1.3.0