NASA's hyperwall wonderwall uses virtual flash SAN

User topics

Article topics

Other stories you might like

ServiceNow ordered a year's worth of hardware to avoid supply chain hassles

CTO shares datacenter secrets with The Reg: NVMe, MariaDB, mid-range x86 CPUs, S3-alike, and more

Simon Sharwood, APAC Editor Wed 25 May 2022 // 04:58 UTC

The tech world's pandemic supply chain meltdown drove ServiceNow to place orders for a year worth of datacenter kit in January 2022, believing that doing so was necessary to get the hardware it needed to cope with growing customer workloads.
"Pre-COVID, I could generally get stuff in 45 days," CTO Pat Casey told The Register at ServiceNow's Knowledge 22 conference in Sydney, Australia, today.
Well-publicized coronavirus-related supply challenges caused ServiceNow's lead time for some networking kit to stretch to 160 days, while servers can take 120 days to arrive.

Continue reading
Marvell CXL roadmap goes all-in on composable infrastructure

Chip biz bets interconnect tech will cement its cloud claim, one day

Tobias Mann Wed 18 May 2022 // 10:28 UTC

Fresh off the heels of Marvell Technology's Tanzanite acquisition, executives speaking at a JP Morgan event this week offered a glimpse at its compute express link (CXL) roadmap.
"This is the next growth factor, not only for Marvell storage, but Marvell as a whole," Dan Christman, EVP of Marvell's storage products group, said.
Introduced in early 2019, CXL is an open interface that piggybacks on PCIe to provide a common, cache-coherent means of connecting CPUs, memory, accelerators, and other peripherals. The technology is seen by many, including Marvell, as the holy grail of composable infrastructure, as it enables memory to be disaggregated from the processor.

Continue reading
Linux kernel patch from Google speeds up server shutdowns

First world problems: Too many NVMe drives, not enough seconds to spare

Liam Proven in Prague Tue 29 Mar 2022 // 17:00 UTC

A new Linux kernel patch from a Google engineer resolves a problem caused by a condition that many of us might quite like to experience – having too many NVMe drives.
The problem is caused by the relatively long time it takes to properly shut down a drive: apparently, as much as four-and-a-half seconds.
Remember Sun's X4500 storage server, originally codenamed Thumper? It was truly radical when it appeared: a 3U dual-processor server, but with a stonking 48 drive bays. These days Google has a bunch of boxes with a still-fairly-impressive 16 NVMe drives attached to each one. And when they have to reboot, they take a long time.

Continue reading
2021 in storage: We waited for a flash price revolution that never came. But about creativity? We can't complain

DPUs in the data centre, object and file get cosy, 3D layer Jenga plus cloud, capacity, and ransomware advances

Chris Mellor Thu 30 Dec 2021 // 12:00 UTC

Analysis While much of the world was in lockdown in 2021, storage boomed. It was a year of ransomware, tech advances, hybrid multi-cloud, a switch to subscriptions and services, hypergrowth in analytics startup funding, and building a DPU data centre makeover.
Suppliers grew, were acquired, struggled, were reborn, and a few (very few) went under. We also saw moves on the high-level exec dancefloor as CEOs and other execs darted between companies, looking to make the optimal career move.

Continue reading
Flash? Nu-uh. Windows 11 users complain of slow NVMe SSD performance

Microsoft aware of the issue months ago, but not fixed yet

Tim Anderson Wed 8 Dec 2021 // 14:41 UTC

Updated Users of Windows 11 are complaining about slow write speeds on NVMe SSD drives, a problem which persists even though it was acknowledged by a Microsoft engineer three months ago.

Continue reading
Seagate demos hard disk drive with an NVMe interface. Yup, one with spinning platters

Oh look, it's the promise of multiple actuators again

Liam Proven in Prague Wed 17 Nov 2021 // 08:04 UTC

At last week's Open Compute Project global summit, Seagate demonstrated a mechanical hard disk drive with an NVMe interface – an interface normally reserved for SSDs. The clue is right there in the name: NVM, Non-Volatile Memory. So the first question is... why?
Well, one purported reason is speed. While Seagate has been promising multi-actuator hard disks for about four years now, you still can't buy them.
The idea is that by having two (or more) separate arms scuttling independently to and fro across the media, hard disks can run fast enough that current SATA interfaces will prove to be a bottleneck. That's 6Gb/s for SATA revision 3, or 600MB/s in reality, while NVMe maxes out at 20Gb/s.

Continue reading
Nvidia says its SmartNICs sizzled to world record storage schlepping status

But did so operating as NICs – the real fun starts in mid-2022 when VMware’s hypervisor for SmartNICs should land

Simon Sharwood, APAC Editor Wed 22 Dec 2021 // 06:02 UTC

Nvidia has claimed a world record for storage IOPS using its Bluefield data processing units (DPUs).
The Bluefield kit is Nvidia’s take on a “SmartNIC” – a network interface card bulked up with a decent CPU so it can run workloads such as firewalls or encryption engines, so that CPUs can be freed for more important tasks. Hyperscalers routinely use SmartNICs and the likes of Nvidia, Intel, and VMware are working to bring them into mainstream data center scenarios.
Nvidia’s benchmark saw a pair of HPE ProLiant DL380 Gen 10 Plus servers hooked up with a dual-100GB ethernet Bluefield aboard each, in a configuration that offered 400Gb/s wire bandwidth between initiator and target. The company used the Storage Performance Development Kit (SPDK) and FIO test, and also targeted the Linux kernel.

Continue reading
Near-instant game loads, richer graphics, low CPU use promised with DirectStorage API coming to Windows PCs

All the gamers NVMe

Tim Anderson Wed 2 Sep 2020 // 17:01 UTC

Microsoft plans to bring the DirectStorage API developed for the forthcoming Xbox Series X to Windows PCs, and a development preview will be available next year.
According to DirectX Senior Program Manager Lead Andrew Yeung, the focus of the new API is PC gaming. Windows storage APIs have become a bottleneck in the path from data on storage volumes to graphics rendered on the GPU.
More detailed graphics means more data to shift, and with game engines breaking down textures into small chunks for memory efficiency, the number of IO requests has increased. NVMe (Non-Volatile Memory Express) storage has excellent I/O performance: 2.4GB/s in Xbox Series X, according to Microsoft's specifications, but the existing PC storage APIs cannot keep up. Taking full advantage of the bandwidth means more than 35,000 IO requests per second.

Continue reading
Green with NVMe: AWS adds more Arm-powered instance types

Local SSD block-level storage comes to EC2 instances powered by home-brew Graviton2 CPUs

Simon Sharwood, APAC Editor Tue 28 Jul 2020 // 04:57 UTC

Amazon Web Services has created three more EC2 instance types that run its home-grown Graviton2 Arm-compatible processors.
The new instances are variants of the M6g, R6g, and C6g instances that the cloud colossus currently offers for general-purpose compute, memory-intensive applications, and compute-intensive workloads, respectively. And the main new feature is the addition of local NVMe-based SSD block-level storage. Adding this support ought to make the three instance types faster for storage-bound workloads. And who doesn’t want faster?
It’s also tough to reject cheaper? AWS claims they “deliver up to 40% better price performance and 50% more NVMe storage GB/vCPU over comparable x86-based instances for a wide variety of workloads.”

Continue reading