Big Memory Solutions

Driven by real-time workloads

Universal Solutions

Unlock the “Persistence” in Intel® Optane™ Persistent Memory Without Modifying Applications

Without Memory Machine software, all apps that needs the persistence capability of PMEM must be modified.

More Memory. Less Cost.

Intel Optane Persistent Memory offer lower cost and higher capacity than DRAM. Memory Machine software takes it to the next level by enabling higher utilization of memory capacity.

Make PMEM Go DRAM Fast

With Memory Machine, you can deploy lower cost and higher capacity PMEM without sacrificing DRAM performance. Memory Machine makes the pool of DRAM and PMEM go DRAM fast.

Create a Highly Available Memory Tier

ZeroIO™ Snapshot technology powers the recovery of terabytes in seconds. This capability transforms the volatile memory tier into a high availability tier.

Application-Specific Solutions

According to IDC, by 2021, 60-70% of the Global 2000 organizations will have at least one mission-critical real-time workload. Below are just a few examples of use cases that are implementing Big Memory now.

Increase Availability with Lightning Fast In-Memory Database Crash Recovery

“Slow” is The State of Art for IMDB Crash Recovery. Typically, a crashed system can be restored from disk or SSD replaying the transaction log to catch up to the last saved state. For one financial services customer of MemVerge this process takes 3 hours to recover 500GB.

As the blast zone for IMDB crashes grows wider and deeper, Intel and MemVerge respond with Big Memory. With Optane persistent memory and Memory Machine software, memory can safely scale-out to petabytes because it’s now possible to recover from crashes in seconds. The state-of the-art for IMDB crash-recovery has been changed from “slow” to “fast”.

Instantly Transmit & Store Massive Amounts of Pub/Sub Market Data with Big Memory

Applications ranging from News, to Risk Analysis, to Trading Engines rely on streaming real-time market data from a network of exchanges. As the number of nodes and messages grows, instability emerges, limiting the scalability of a pub/sub networks.

The Memory Machine software Publish/Subscribe messaging API provides MemVerge customers with the tools to create a Pub/Sub system to instantly transmit and store massive amounts of time sequence data such as stock exchange trading ticks.

Scale Memory Capacity and Availability for VMware ESX

Host is not responsive. Virtual machines cannot be started. Applications in virtual machines run out of memory. These are all symptoms of a problem called “DGM,” Data-is-Greater-than-Memory. The capacity, availability, and crash recovery challenge is even greater when large data sets, and real-time analytics are involved.

Big Memory addresses VMware’s need for more memory with support for higher-density / lower-cost persistent memory that allows more memory to be configured in a single physical server. Memory Machine software then takes capacity to the next level with the ability to scale-out memory in a virtual server cluster to form massive memory lakes. Memory Machine protects all that data with memory data services including snapshot, replication and lightning-fast recovery.

Greater Memory Capacity & Plug-Compatible Access to Persistent Memory for ML Applications

Inference from a pretrained model often requires large databases that must reside in system memory to achieve sufficient transactional performance. When resources become tight, nominally less used memory items are written to disk to make more space, only to be swapped back later. This back-and-forth has severe impacts, drastically reducing the performance.

Facebook Deep Learning Recommendation Model (DLRM), Image Recognition, and Stanford GraphSAGE are three ML applications that Memory Machine software has powered to large persistent memory configurations eliminate the performance penalty of memory swapping.

Share This