Presenting Mangstor's NVMe superfast flash storage pocket rocket

In investment terms, it's the most bang the industry could get for $4m


Comment The Register storage desk thinks NVMe fabric linking for storage arrays will be very big, as it's a SAN/NAS latency killer. Startup Mangstor has built an NVMe fabric-accessed array, so we've seen what such a beast looks like.

We looked at its MX6300 PCIe flash card recently and thought it was a marvelously fast product. We wrote that it has custom hardware, with three FPGAs, a 100-core, 500MHz processor, firmware, 4GB of Micron DDR3 RAM and some ST-MRAM.

Mangstor has built an NX-Series NVMe-over-Fabric enterprise all-flash storage array, which won the Best-of-Show award for “Most Innovative Flash Technology” during August's Flash Memory Summit. It uses MX6300 cards inside a 2U x86 server running a Titan software stack.

This array supports both NVMe over Ethernet and InfiniBand fabrics packaged through an RDMA cluster scale-out architecture. The Ethernet connection uses RDMA over Converged Ethernet (RoCE) fabrics or InfiniBand to connect users to its s NX6320-Series array.

MX6300_Performance

MX6300 performance data

Mangstor’s Best-of-Show award statement said it provides "lower latency and higher IOPS performance than traditional SAN solutions”. Well, we would certainly hope so, but the real interesting comparison would be against networked all-flash arrays and not traditional SAN systems.

The performance is host data access "at nearly identical latencies as if they were accessing local PCIe SSDs; increasing application performance and providing over 3 million random 4KB read and 2.25 million 4KB write IOPS" with latencies under 200ms.

NX6320_Performance

NX6320 performance table

An individual MX6300 flash card has 90/15 microsecond read/write latency while the NX6320 array performs with 110/30 microsecond read/write latency; not much of an increase at all. We’re told this is actually lower latency than typically seen on most NVMe SSDs from inside a server. It’s an order of magnitude or more better than typical FC or iSCSI latencies.

Next page: Expansion prospects

Similar topics

Broader topics


Other stories you might like

  • 381,000-plus Kubernetes API servers 'exposed to internet'
    Firewall isn't a made-up word from the Hackers movie, people

    A large number of servers running the Kubernetes API have been left exposed to the internet, which is not great: they're potentially vulnerable to abuse.

    Nonprofit security organization The Shadowserver Foundation recently scanned 454,729 systems hosting the popular open-source platform for managing and orchestrating containers, finding that more than 381,645 – or about 84 percent – are accessible via the internet to varying degrees thus providing a cracked door into a corporate network.

    "While this does not mean that these instances are fully open or vulnerable to an attack, it is likely that this level of access was not intended and these instances are an unnecessarily exposed attack surface," Shadowserver's team stressed in a write-up. "They also allow for information leakage on version and build."

    Continue reading
  • A peek into Gigabyte's GPU Arm for AI, HPC shops
    High-performance platform choices are going beyond the ubiquitous x86 standard

    Arm-based servers continue to gain momentum with Gigabyte Technology introducing a system based on Ampere's Altra processors paired with Nvidia A100 GPUs, aimed at demanding workloads such as AI training and high-performance compute (HPC) applications.

    The G492-PD0 runs either an Ampere Altra or Altra Max processor, the latter delivering 128 64-bit cores that are compatible with the Armv8.2 architecture.

    It supports 16 DDR4 DIMM slots, which would be enough space for up to 4TB of memory if all slots were filled with 256GB memory modules. The chassis also has space for no fewer than eight Nvidia A100 GPUs, which would make for a costly but very powerful system for those workloads that benefit from GPU acceleration.

    Continue reading
  • GitLab version 15 goes big on visibility and observability
    GitOps fans can take a spin on the free tier for pull-based deployment

    One-stop DevOps shop GitLab has announced version 15 of its platform, hot on the heels of pull-based GitOps turning up on the platform's free tier.

    Version 15.0 marks the arrival of GitLab's next major iteration and attention this time around has turned to visibility and observability – hardly surprising considering the acquisition of OpsTrace as 2021 drew to a close, as well as workflow automation, security and compliance.

    GitLab puts out monthly releases –  hitting 15.1 on June 22 –  and we spoke to the company's senior director of Product, Kenny Johnston, at the recent Kubecon EU event, about what will be added to version 15 as time goes by. During a chat with the company's senior director of Product, Kenny Johnston, at the recent Kubecon EU event, The Register was told that this was more where dollars were being invested into the product.

    Continue reading

Biting the hand that feeds IT © 1998–2022