Home Blog Page 344

WekaIO links up with Nvidia GPU Direct to uncork AI I/O bottlenecks

WekaIO has devised a “production-ready” framework to help artificial intelligence installations to speed up their storage data transfers. The basic deal is that WekaIO supports Nvidia’s GPUDirect storage with its NVMe file storage. Weka says its solution can deliver 73 GB/sec of bandwidth to a single GPU client.

The Weka AI framework omprises customisable reference architectures and software development kits, centred on Nvidia GPUs, Mellanox networking, Supermicro servers (other server and storage hardware vendors are also supported) and Weka Matrix parallel file system software.

Paresh Kharya, director of product management for accelerated computing at Nvidia, provided a quote: “End-to-end application performance for AI requires feeding high-performance Nvidia GPUs with a high-throughput data pipeline. Weka AI leverages GPUDirect storage to provide a direct path between storage and GPUs, eliminating I/O bottlenecks for data intensive AI applications.”

Nvidia and Weka say AI data pipelines have a sequence of stages with distinct storage IO requirements: massive bandwidth for ingest and training; mixed read/write handling for extract, transform, and load (ETL); and low latency for inference. They say a single namespace is needed for entire data pipeline visibility. 

Weka and Nvidia

Weka AI bundles are immediately available from Weka’s channel. You can check out a Weka AI white paper (registration required.)

Pure and Cohesity team up with FlashRecover data protection

Pure Storage is reselling Cohesity software with its FlashBlade storage array to provide a single flash-to-flash-cloud, data protection and secondary data management system.

Mohit Aron

Mohit Aron, Cohesity CEO and founder, said in a statement: “We are thrilled to partner with Pure in bringing to market a solution that integrates exceptional all-flash capabilities and cutting-edge data protection offerings that together unleash new opportunities for customers.

Called Pure FlashRecover, the team effort combines the FlashBlade array with a white box server that runs Cohesity’s Data Platform software. This is not an appliance in the sense that it is a dedicated and purpose-built product but it can be used in an appliance-like manner. Pure FlashRecover is a jointly-engineered system, with disaggregated and independently scalable compute and storage resources. The FlashBlade array can perform functions beyond providing storage for Cohesity software.

Flash-to-flash-to-cloud. FlashRecover is the deduping flash store in the diagram

FlashRecover can function as a general data protection facility for Pure Storage and other suppliers’ physical, virtual, and cloud-native environments, with faster-than-disk restore and throughput from the all-flash FlashBlade array. Most functionality of the hyperconverged, scale-out Cohesity DataPlatform is also available to customers. Features include tiering off data to a public cloud, ransomware protection, copy data management, data supply to analytics and test and dev.

Integration

Pure has become a Cohesity Technology Partner and the two companies have integrated their environments. Cohesity Helios management software auto-discovers FlashBlade systems and Cohesity uses FlashBlade snapshots.

Cohesity spreads the data across available space on FlashBlade to maximise restore performance and enhance efficiency. The software is optimised to provide performance even when the storage for the data is from disaggregated FlashBlades.

FlashRecover will be sold by Pure’s channel and supported by Pure. Cohesity and Pure are looking forward to further joint technology developments from this point.

ObjectEngine replacement

Last month, Pure canned the FlashBlade-based ObjectEngine backup appliance. The company told us it was “working with select data protection partners, which we see as a more cohesive path to enhancing those solutions with native high performance and cloud-connected fast file and object storage to satisfy the needs in the market.” Now we see that Cohesity replaces the ObjectEngine software and FlashRecover replaces the ObjectEngine appliance.

Pure FlashRecover, Powered by Cohesity is being tested by joint customers today and will be generally available in the United States in the fourth quarter, and elsewhere at unspecified later dates. Proof of concepts are available now for select customers .

Spin Memory spins benefits of Universal Selector

Spin Memory has designed a ‘Universal Selector’ transistor that improves DRAM array density by 20 to 35 per cent, according to the company. It says the technology increases MRAM density by up to five times and accelerates MRAM operations.

Spin Memory derives its name from the field it works in. The company is a Spin Transfer Torque MRAM developer, which positions MRAM as a replacement memory technology for CPU caches using Static RAM. It claims its STT-MRAM could function as storage-class memory like Intel’s Optane.

Tom Sparkman

Tom Sparkman, CEO of Spin Memory, provided an announcement quote: “Our latest breakthrough innovation allows for exciting new advancements and capabilities for [advanced memory] technologies – in addition to pushing MRAM into the mainstream market.” 

Spin Memory’s Universal Selector is, according to the announcement, a selective, vertical epitaxial cell transistor with a channel that is electrically isolated from the silicon substrate. Epitaxial refers to the growth of a crystal in a particular orientation on top of another crystal whose orientation determines the growth direction of the upper crystal. A vertical epitaxial cell is grown vertically above the underlying crystal. Epitaxy is used in semiconductor manufacturing to form  layers and wells.

Spin Memory Universal Selector patent diagram (fig 18). Key; 602 – semiconductor substrate, 604 – diffusion lines, 1102 – word lines, 1304 – gate dielectric layer, 1602 – epitaxially grown monocrystalline semiconductor material.

Spin Memory thinks this technology can be applied outside the MRAM field to DRAM, ReRAM and PCRAM, for example, hence the ‘universal’ attribute.

The electrical isolation prevents DRAM row hammer attacks, Spin Memory says.These are malware instances where repetitive (hammering) memory access patterns are made to to DDR3 and DDR4 cells, causing leakage and so altering cell contents in adjacent cell rows.

Charlie Slayman, IRPS 2020 technical program chair, was quoted in Spin Memory’s announcement: “Spin Memory’s Universal Selector offers a novel way to design vertical cell transistors and has been presented to the JEDEC task group evaluating solutions to the row hammering problem.”

Spin Memory claims Universal Selector gives any developer of non-Flash memories the means to “drastically improve density for almost every memory technology on the market without requiring an investment in specialised hardware or resources.”

Jim Handy of Object Analysis, an analyst firm, said: “This is a very clever use of the vertical transistors that are the basis of 3D NAND flash. If a vertical transistor can make NAND smaller, then why not harness it for other technologies?”

“Selectors have always been a vexing issue for emerging memories.  A number of developers focus all of their effort on the bit element but none on the selector. Both are significant challenges and both need to be solved or you have nothing.”

Spin Memory has patented the Universal Selector – US 2020/0127052 AI. It will share additional technical details at the virtual 31st Magnetic Recording Conference on Thursday, August 20 at 10:40 a.m. PDT.

IBM storage boss Ed Walsh surfaces at ChaosSearch as CEO

Ed Walsh left his job as head of IBM Storage last week to take on the CEO role at ChaosSearch, a log data analytics startup. Co-founder Les Yetton has stepped aside to make way for Walsh.

ChaosSearch CTO Thomas Hazel told us in a briefing this week that Walsh “has the experience and vision to enable us to very rapidly scale to meet customer demand we are seeing”.

Hazel and Yetton set up the company in 2016 to devise a faster, more-efficient way of searching the incoming flood of log and similar unstructured data heading for data lakes. They saw that analysis was becoming IO-bound because the Moore’s Law progression of compute power scaling was slowing. Their answer was to build software toc ompress and accelerate the analytics IO pipeline.

Thomas Hazel

Walsh said in a press statement today: “The decision to leave IBM was extremely difficult for me, but the decision to join ChaosSearch was very easy. Once I saw the unique and innovative approach they take to enable ‘insights at scale’ within the client’s own Cloud Data Lake and slash costs, I knew I wanted to be part of this company.”

He added: “Unlike the myriad log analytic services or roll-your-own solutions based upon the Elastic/ELK stack, ChaosSearch has a ground-breaking, patent-pending data lake engine for scalable log analytics that provides breakthroughs in scale, cost and management overhead, which directly address the limits of traditional ELK stack deployments but fully supports the Elastic API.”

The ELK stack refers to ElasticSearch, Logstash and Kibana; three open-source products maintained by Elastic and used in data analysis.

Background 

ChaosSearch saw that data in data lakes needed compressing for space reasons, and also extracting, transforming and loading (ETL) into searchable subsets for Elastic Search and data warehouses such as Snowflake. This involved indexing the data to make it searchable. What the founders invented was a way of compressing such data and indexing it in a single step.

Hazel told us Gzip compression takes a 1GB CSV file and reduces it to 500MB. Paquet would turn it into an 800MB file while Lucene would enlarge it three to five times by adding various indices. ChaosSearch reduces the 1GB CSV file to 300MB, even when indexing is performed.

Bump the 1GB CSV file to multiple TB or PB levels and the storage capacity savings become quite significant. Overall analytics time is also greatly reduced because the two separate compression and ETL phases are condensed into one.

The compression means that more data can be stored in the analysis system and this improving analytics results.

Ed Walsh.

ChaosSearch thinks a cloud data lake can be stored in cloud object storage (S3) and function as a single silo for business intelligence and log analytics. The company is initially focusing on the log analytics market and says its technology can run search and analytics directly on cloud object storage. Walsh claimed ChaosSearch technology is up to 80 per cent less expensive than alternatives, in our briefing with the company.

The ChaosSearch scale-out software will be available on Amazon’s cloud initially and be extended to support Azure and GCP. Upstream analysis tools such as Elasticsearch access its data through standard APIs.

Check out a white paper for a detailed look inside ChaosSearch’s technology.

Quobyte storage software takes a bite at HPC and machine learning

Profile Quobyte produces unified block, file and object storage for high-performance computing (HPC) and machine learning.

Its software runs on-premises and in the Google cloud and is used for data analytics, machine-learning, high-frequency trading, life sciences and similar petabyte-scale, fast-response applications. The company also has a TensorFlow plug-in.

Quobyte was established by CEO Björn Kolbeck and CTO Felix Hupfeld in Berlin in 2013. The pair studied computer science and did distributed storage systems research at Berlin’s Zuse Institute, where they developed the open source XtreemFS distributed file system as part of their PhD. They then worked for Google and this gave them the idea that they could build a Google-like scalable, distributed, parallel access and multi-protocol storage system for HPC customers.

Kolbeck and Hupfel thought that existing HPC storage was placed in restrictive silos such as file or object, was hardware-dependent, and needed a lot of administration. Their response was to develop high-performance and multi-protocol storage software that was independent of its hardware supplier and manageable by a small team.



Björn Kolbeck (left) and Felix Hupfeld

The software use commodity server hardware and is intended to replace several separate storage vaults while delivering good random and small file IOPS and large file throughput.

Competition

Quobyte is a small company. To give an indication of size, the company lists a dozen customers on its website and has 19 employees, according to LinkedIn profiles. Alstin Capital, an investor, valued Quobyte at $35m in 2018 and describes it as an early growth company. The funding history is a little opaque but the last capital raise was in 2016. According to one source, the company is exploring a fresh round.

As an HPC storage software supplier, Quobyte competes with Ceph, DDN, Intel DAOS, Lustre, Panasas, Qumulo, IBM’s Spectrum Scale and WekaIO, amongst others. Compared to most of these, Quobyte is a minnow without substantial backing.

Open source Ceph, which also supplies file, block and object access access storage, has an underlying object store, with file and block access layered above that. Kolbeck and Hupfeld thought they would get better performance from a file system base with block and object access layered on it.

IBM’s Spectrum Scale has no block storage and needs fairly complex tuning and administration. Panasas PanFS software has recently become independent of its hardware, and has high-performance but this is limited to file access. DDN HPC storage requires DDN hardware.

Quobyte technology

Quobyte has developed a Data Centre File System or Unified Storage Plane – two names for the same thing – which supports POSIX, NFS, S3, SMB and Hadoop file and block access methods.

The software has a distributed parallel, fault-tolerant, highly scalable and POSIX-compliant file system at its heart. This links Quobyte nodes (servers) with Quobyte clients in accessing servers, which make remote procedure calls (RPCs) to the Quobyte servers. 

The servers are nodes in a Quobyte storage cluster. Each node’s local file system handles block mapping duties using SSDs and HDDs. The system uses a key:value store designed for file system use and metadata operations are not part of the IO path.

Quobyte software can host any number of file system volumes which can be assigned to tenants. Various system policies control data placement, tiering, workload isolation, and use of partitions. The software has built-in accounting, analytics and monitoring tools plus built-in hardware management. It also supports cloud-native workloads with Docker, Mesos and Kubernetes and supports OpenStack and a Google TensorFlow plug-in.

Quobyte’s software operates in Linux user mode and not privileged kernel mode. This avoids context switches into kernel mode and means the software is safer as it can’t execute potentially damaging kernel mode instructions.

Three basic services

The software has three basic services; Registry Service to store cluster configuration details, Metadata Data Service for file access data and Data Service for actual file IO. One Data Service operates per node and can handle hundreds of drives. There can be hundreds or more data service nodes.

The metadata service is deployed on at least four nodes and can scale to hundreds of nodes. Placing metadata on NVMe SSDs provides the lowest latency accesses.

A client can issue parallel reads and writes to the servers. All clients can access the same file simultaneously. Byte-range or file-level locking is supported across all protocols that support locks to prevent corruption from multipole writes to the same data.

Get more information from a Quobyte white paper.

Nutanix Clusters takes on-premises Nutanix to AWS

Nutanix is ready to announce Nutanix Clusters. This brings the on-premises Nutanix experience to AWS and opens another front in the company’s battle with VMware.

Sources close to the company say Nutanix Clusters in AWS (NCA) has been in an early-access test phase for many months and is now robust and ready to move into general availability.

NCA runs on bare metal all-flash servers in AWS and uses AWS networking. Customers spin up servers using their AWS account and deploy Nutanix software on them. This process uses AWS Cloud Foundation, Amazon’s facility to provision and model third-party applications in AWS. On-premises Nutanix licenses can be moved to AWS to instantiate NCA there.

VMware uses its ESX hypervisor as an overlay atop AWS networking and this can sap resources and become a performance bottleneck, according to Nutanix sources.

NCA supports four types of AWS instance, including Large CPU, Large Memory and Large Storage. The Nutanix Prism management console can be used to manage NCA.

NCA bursts on-premises Nutanix deployments to AWS to cope with spikes – for example, an immediate requirement to add 1,000 virtual desktops. It also has disaster recovery capabilities. 

Customers can use NCA to migrate on-premises applications running on Nutanix to AWS. There is no need to re-engineer applications as the software runs the Nutanix environment transparently across the public cloud and on-premises worlds. 

When no longer needed, a Nutanix cluster in AWS can be spun down with its data stored in S3, incurring S3 charges only, until spun up again. The the spun-up NCA is rehydrated from the S3 store. We understand this go-to-sleep facility will follow the main NCA release in a few months.

A blog by Nutanix CTO Binny Hill provides more background on NCA and there is more info an early-access solution brief.

Hubstor offers unified cloud platform for backup and archiving

Cloud Love

Sponsored Tucked away in his files, HubStor CEO Geoff Bourgeois still has the scribbled notes in which he and co-founder Greg Campbell first sketched out the idea for their business back in 2015. He recently looked those notes out, he says, and was gratified to see that their vision hasn’t changed that much. 

“There may have been some twists and turns along the way,” he says, “but my main impression was, ‘Wow. We’ve really stuck to plan.’”

That plan, from the start, was for the Kanata, Ontario-based business to build a unified platform that enables companies to tap into the scalability and cost-efficiency of the cloud to meet their backup and archiving needs.

Along the way, HubStor’s promise to help them protect their data, simplify their IT infrastructure and achieve compliance, all via a software-as-a-service (SaaS) offering, has won over many corporate customers, including a handful from the Fortune 100.

“While we bootstrapped the company ourselves in its early days,” Bourgeois reports, “the business is now customer-funded by a recurring revenue model and seeing healthy organic growth.”

A gap in the market

The key to this success, as with so many businesses, came from spotting a gap in the market. Back in 2015, Bourgeois says, there was a great deal of interest among IT decision-makers in using the cloud for backup and archiving. Many were looking to escape the costly and time-consuming trap of having to provision and manage more and more on-premise infrastructure to meet their storage needs. But when they looked into the cloud, they often complained, their options were limited to commodity infrastructure-as-a-service (IaaS) offerings.

These presented significant challenges. While IaaS might provide a bare-bones cloud location in which to store data, these services didn’t connect easily to the workloads or applications that organisations were running; nor did they help IT admins to apply data management rules and policies around backup and archiving to the data sets they were looking to migrate. As Bourgeois puts it: “There was no easy on-ramp to the cloud.”

“That’s when a lightbulb came on for us,” he continues. “There was a big disconnect here – and what was missing was a SaaS layer that would make the cloud more consumable for backup and archiving.”

The value of a ‘bridge’ connecting company data sets to low-cost, scalable cloud storage, in a logical, easy-to-manage way, has only increased over the intervening five years, according to Bourgeois.

The HubStor interface gives IT administrators a single console via which they can view all backup and archive workloads, handle each one according to its particular data-management needs, and migrate it to the cloud region of their choice, or to multiple regions where appropriate.

They can allocate data to different storage tiers in the cloud, in order to get the most from their storage budget, and apply special rules for specific enterprise use cases, such as security, regulatory compliance and litigation preparedness.

The data involved may be structured or unstructured, database or blob. It might originate from on-premise systems, or from other private or public clouds. For example, the back-up of corporate SaaS applications, including Microsoft 365, Slack and Box, is now a high-growth revenue stream for the business, says Bourgeois.

Twists and turns

The rapid growth of back-up for SaaS applications is just one of those “twists and turns” in the company’s history and an area where it continues to expand. Backup for files stored in Box, a cloud content management provider, for example, was added in November 2019.

A more recent launch is backup for virtual machines (VMs); until as recently as last year, even Bourgeois doubted there might be sufficient demand for a new product in this market, given that Commvault, Veeam and Veritas are all active and established players there.

“I thought those guys had this market sewn up, but we’ve been asked again and again at trade shows whether we offer backup for virtual environments. And the people asking would say they were customers of those other companies, but looking for something different, something more cloud-based, something more modern in terms of user experience. So, some time around the end of 2019 we were like, ‘OK, it’s time for us to do this.’”

In May 2020, HubStor announced the release of its BaaS (Backup-as-a-service) solution for VMware vSphere environments. Customers can use either a dedicated or shared BaaS model, with pricing based mainly on their cloud storage consumption, rather than the number of VMs protected, as is more typically the case with competitors. HubStor also plans to support other virtual infrastructures, including those based on Microsoft Hyper-V and Nutanix, in the near future.

Customer expansion opportunities

This new offering could be a valuable way to attract new customers, who might then go on to explore HubStor’s other offerings: file system tiering, Office 365 backup, journaling or content archiving, for example.

It makes a lot of sense to customers to do this and is a regularly observed trend, says Bourgeois. “It’s a win-win for customers when they throw multiple workloads at us. I mean, it’s fine if a customer signs up to HubStor and just uses it for one particular workload or one particular use case. But there’s a very small, incremental uptick in cost to them when they want to leverage it for other things, and so they start to achieve very significant economies of scale quite quickly with the compute investment they’ve made to use our SaaS.”

A great example, he says, is Los Angeles-based Virgin Hyperloop One, which is working to make this thoroughly modern form of mass transportation a reality, using electric propulsion and electromagnetic levitation in near-vacuum conditions. It may not be HubStor’s largest corporate customer from a revenue perspective, says Bourgeois, but it’s one that is taking advantage of close to the full scope of its platform’s capabilities.

“So for example, Virgin Hyperloop One is backing up SaaS applications, including Microsoft 365, and doing long-term retention of audit log data from SaaS applications. They’re doing virtual machine back-up. They’re tiering file systems and they’re leveraging HubStor for legal discovery and security audits, too,” he says.

Conversely, existing HubStor customers may be keen to adopt the new VM backup product. Among them is Axon Corporation, a Scottsdale, Arizona-based industrial machinery company. According to Axon’s manager of infrastructure systems Ryan Reichart, the company is looking forward to replacing its existing backup product, which he describes as “expensive and extraordinarily complex.”

Says Reichart: “One exciting feature is that [Hubstor’s VM backup] integrates directly with the SaaS management dashboard and storage accounts of our other HubStor products in use – file tiering and Office 365 backup.”

In other words, HubStor will bring new cost-efficiency and simplicity to backup and archiving that Axon has previously been lacking.

A new simplicity

It’s a similar story at other customers, says Bourgeois, and simplicity has been a mantra for him and the HubStor team from the start. With a user experience refresh planned for this year, the team will be looking to make HubStor’s SaaS interface even easier to use.

At its heart, a SaaS platform “gets customers out of the weeds”, he says. They no longer have to manage on-premise tape libraries or backup servers or deal with licensing issues. Above all, they get a true-to-cloud consumption pricing model, typically for the first time.

“With HubStor in place, they can see exactly what underlying cloud resources they’re consuming, and the costs associated with that, plus a simple subscription fee on top for using our SaaS, which is a totally transparent mark-up on the use of cloud resources,” he explains.

This means that customers can move storage around, to lower cost storage tiers. For example, if a cloud provider introduces sharper pricing tactics, they can reassess their approach and shrink the underlying cloud costs still further. “In some cases,” Bourgeois concludes, “we’ve been able to help customers achieve a million-dollar a year budget relief from escaping some of the legacy approaches they were using. That’s a good feeling and it’s a big source of pride to us.”

This article was sponsored by Hubstor.

Intel regains IO500 HPC bragging rights from WekaIO

Intel has leapfrogged WekaIO to claim the IO500 fastest HPC file system. WekaIO supplanted Intel to take top place in the previous IO500 league table, which was published in November 2019.

Compiled by The Virtual Institute for IO, the IO500 benchmarks measures bandwidth, metadata and namespace searching. Last November, WekaIO scored 938.95 versus Intel’s 933.64. In the latest ranking, Intel scored 1,792.98 with WekaIO hitting the same number as last time. Here are the top five results:

Intel also took the IO500 third slot with a Texas Advanced Computing Centre Frontera system using DAOS, which scored 768.80.

DAOS (Distributed Application Object Storage) is Intel’s open source file system for high performance computing. It accelerate file accesses using NVRAM and Optane storage-class memory plus NVMe storage in user space. 

The DAOS software stack is still under development – for example, Intel has adapted the IO driver to work better with the IOR and MDTEST components of the IO500 benchmark.

DAOS placed second on the November 2019 ‘Full List’, using 26 client nodes. Better performance can be expected with a larger set of client nodes, particularly for metadata tests that scale with the number of client nodes. In the latest test, Intel doubled node count to 52 and nearly doubled performance.

Over to you, Weka.

This Week in Storage, featuring Qumulo, Actifio and more

This week, AWS has made Qumulo filer software available in the AWS government cloud; Actifio backs up SAP HANA to object storage in GCP; and LucidLink is doing cloud collab stuff with Adobe Premiere Pro.

Qumulo AWS takes on Azure NetApp Files

Scalable file system supplier Qumulo has announced its availability in the AWS GovCloud (US) through the AWS Marketplace.

Qumulo says Government organisations can now integrate their file data with legacy applications in private cloudand cloud-native applications in AWS GovCloud (US) with a single file data platform.

The company is working with Corsec Security Inc. to gain various US government certifications for its software. The company said it aims to make Qumulo the strategic choice for all types of Controlled Unclassified Information (CUI) and unclassified file data., as well as the upcoming FIPS 140-2 and Common Criteria EAL2+ certifications of its platform.

NetApp, a Qumulo competitor, this week announced Azure NetApp Files is in the Azure government cloud

Actifio HANA DR costs 86 per cent less on GCP

Copy data manager Actifio is waving a tech validation report from ESG that says it reduced backup and disaster recovery (DR) infrastructure costs by 86 per cent when protecting SAP HANA workloads with Google Cloud Platform (GCP) object storage. The comparison is with legacy backup approaches using high-performance block storage. 

ESG found the same high levels of performance from a DR copy running off Google Nearline object storage as their production instances running on Google Persistent disk block storage.

ESG Senior Validation Analyst Tony Palmer said: “Cloud object storage is typically 10x inexpensive than the cloud SSD/flash block storage. Actifio’s ability to recover SAP HANA database in just minutes from cloud object storage, while delivering the I/O performance of an SSD/flash block storage is very unique in the industry and reduces cloud infrastructure costs by more than 80 per cent for enterprises.” 

You can download the ESG Actifio SAP HANA Technology Review.

LucidLink builds cloudy Adobe file construction

LucidLink, which supplies accelerated cloud-native file access software, is partnering with Adobe Premiere Pro so its users can edit projects directly from the cloud.

Generally, Adobe Premiere Pro video editing software users edit local files because access is fast. However, team working  and remote team working require multi-user access to remote files. LucidLink’s FileSpaces can provide teams with on-demand access to media assets in the cloud that are accessed as if they were on a local drive.

Sue Skidmore, head of partner relations for Adobe Video, said “With so many creative teams working remotely, the ability to edit Premiere Pro projects directly from the cloud has become even more important. We don’t want location to hold back creativity. Now Premiere users can collaborate no matter where they are.”

Filespaces provides a centralised repository with unlimited access to media assets from any point in existing workflows. The pandemic has encouraged remote working. Peter Thompson, LucidLink co-founder and CEO, provided a second canned quote: ”Our customers report they can implement workflows previously considered impossible. We are providing the missing link in cloud workflows with ‘streaming files.’”

Shorts

Actifio has announced technical validation and support for Oracle Cloud VMware Solution (OCVS), Oracle’s new dedicated, cloud-native VMware-based environment. OCVS enables enterprises to move their production VMware workloads to Oracle Cloud Infrastructure, with the identical experience in the cloud as in on-premises data centres. It integrates with Oracle’s second-generation cloud infrastructure. OCVS is available now in all public regions and in customer Dedicated Region cloud instances.

Taiwan-based Chenbro has announced the RB23712, a Level 6, 2U rackmount server barebone (no CPUs, fitted drives) with 12 drive bays designed for storage-focused applications in the Data Center and HPC Enterprise. It pre-integrates an Intel Server Board S2600WFTR with support for up to two, 2nd GenerationXeon Scalable Processors. The RB23712 offers Apache Pass, IPMI 2.0 and Redfish compliance, and includes Intel RSTe/Intel VROC options.

Microchip Technology has introduced the latest member of the Flashtec family, the Flashtec NVMe 3108 PCIe Gen 4 enterprise SSD controller with 8 channels. It complements the 16-channel Flashtec NVMe 3016 and provides a full suite of PCIe Gen 4 NVMe SSD functions. The 3108 is intended for use by M.2 and the SNIA Enterprise and Data Center SSD Form Factor (EDSFF) E1.S drives.

Microchip 3108

Nutanix says it has passed 2,500 customers for Nutanix Files. Files is part of a Nutanix suite for structured and unstructured data management, which includes Nutanix Objects, delivering S3-compatible object storage, and Nutanix Volumes for scale-out block storage.

Penguin Computing has become a High Performance Computing (HPC) sector reseller and solution provider of Pavilion Hyperparallel Flash Arrays (HFA).

Quantum has announced its ActiveScale S3-compatible object store software has been verified as a Veeam Ready Object Solution.

Synology has launched new all-flash storage and a line of enterprise SSDs. The FS3600 storage system is the newest member of Synology’s expanding FlashStation family of network-attached storage (NAS) servers. Synology has also announced the release of SATA 5200 SATA SSDs and SNV3400 and SNV3500 NVMe SSDs.

The FS3600 features a 12-core Xeon, up to 72 drives, and 56GbitE support. The new SSDS can fit in its enclosure and have 5-year warranties. They integrate with Synology DiskStation Manager (DSM) for lifespan prediction based on actual workloads.

Data replicator and migrator WANdisco said it is the first independent software vendor to achieve AWS Competency Status in data migration.

Zerto is reprinting a short Gartner report: “What I&O leaders need to know about Disaster Recovery to the cloud.” The report assumes that by 2023,” at least 50 per cent of commodity-server workloads still hosted in the data centre will use public cloud for disaster recovery.” It’s an eight-minute read and you can get it, with minimal registration.

IBM storage boss Ed Walsh leaves the company

Ed Walsh, general manager of IBM’s storage business, is leaving the company, according to sources who say he is not joining a competitor.

An IBM spokesperson said: “Yes, we can confirm Ed Walsh is leaving IBM and we wish him well at his new opportunity.”

Ed Walsh

Walsh was hired to run IBM’s storage division in July 2015. He joined the company from Catalogic, a copy data management company, where he had the CEO and president roles.

Walsh has also worked for IBM before, via the company’s 2010 acquisition of Storwize, a storage array supplier, where he was also CEO. 

IBM appointed a new CEO, Arvind Krishna, in April this year. The storage business within IBM recently recorded its third quarter of revenue growth, helped by z15 mainframe storage sales.

Western Digital loves flash more than hard drives

Western Digital is generating more revenue from flash and SSDs than from disk drives. Does this show that WD is losing out big time to Seagate in the key enterprise nearline HDD market? Or is it a forever thing?

It appears to be the latter, judging from CEO David Goekeler’s comments on the Q4 earnings call yesterday. “We believe flash is the greatest long-term growth opportunity for Western Digital and is an area where we’ve already had a tremendous foundation with consumer cards, USB drives and client and enterprise SSDs.”

Our sister publication The Register has covered WD’s Q4 and full fiscal 2020 year results and we’ve dived into the numbers to see how disks and SSDs fared relative to each other.

WD shipped 23.1 million disk drives in the quarter, down 16.6 per cent from 27.7 million shipped a year ago. HDD revenue was $2.05bn, down 3.7 per cent on the year ago’s $2.13bn. Reflecting capacity price trends, total HDD exabytes shipped – about 110 EB – was 12 per cent on the year, according to Wells Fargo senior analyst Aaron Rakers.

He told subscribers: “WD’s Nearline HDD capacity shipments grew ~30 per cent y/y, or leaving us to estimate flat q/q. We estimate nearline HDD capacity shipped at ~76 EBs, which compares to Seagate shipping 79.5 EB (+128 per cent y/y), and Toshiba shipping 8.4 EB (-14 per cent y/y) in the June quarter. We estimate WD’s non-nearline HDD capacity shipped at ~33 EBs, or -15 per cent y/y and -5 per cent q/q.“

So it looks like WD was not a beneficiary of Toshiba’s calamitous quarter i hard drives, which saw the company take a big hit from an outage at its Philippines disk drive plant.

Western Digital Q4 fy2020 flash and HDD revenue chart; two diverging curves.

WD says 14TB sales are doing well, but the HDD business has not been helped by Seagate’s dominance of the 16TB nearline disk business. WD was late to market with 16TB and 18TB drives and aims to remedy this by ramping 18TB drive production to one million drives in the next quarter.

WD expects cloud nearline disk drive buyers to slow purchases this quarter: “We feel like we’re definitely going into a digestion phase,” Goekeler said. “If we look at — we’re coming off of three really strong quarters of exabyte shipment and the demand signals we’re getting are going to be – are a little bit down for the next quarter or two.”

Let’s now take a quick peek at WD’s flash / SSD business, where revenue of $2.24bn was 48.6 per cent higher than the year ago quarter ($1.51bn).

Goekeler noted “healthy demand for our flash-based notebook solutions drove record revenue in our OEM end market,” in the earnings call, and added: “Enterprise SSD revenue in the quarter grew nearly 70 per cent sequentially and our revenue share increased to the low double digits. This will remain an important area of focus within our flash portfolio.”

To conclude, the disk drive business is now secondary to flash, and NAND wafers drive more revenue for WD than disk drive platters.

Fungible Inc: Why you need DPUs for data-centric processing

A flurry of add-on processor startups think there is a gap between X86 CPUs and GPUs that DPUs (data processing units) can fill.

Advocates claim DPUs boost application performance by offloading storage, networking and other dedicated tasks from the general purpose X86 processor.

Suppliers such as ScaleFlux, NGD, Eideticom and Nyriad are building computational storage systems – in essence, drives with on-board processors. Pensando has developed a networking and storage processing chip which is used by NetApp and Fungible is building a composable security and storage processing device.

Pradeep Sindhu.

A processing gap has opened up, according to Pradeep Sindhu, Fungible co-founder and CEO. GPUs have demonstrated that certain specialised processing tasks can be better carried out by dedicated hardware which is many times faster than an X86 CPU. The idea of Smart NICs, host bus adapters and Ethernet NICs with on-board, offload processors for TCP/IP offload and data plane acceleration, have educated us all that X86 servers can do with help in handling certain workloads.

In a press briefing this week, Sindhu outlined a four-pronged general data centre infrastructure problem.

  1. Moore’s Law is slowing and the X86 speed improvement rate is slowing while data centre workloads are increasing.
  2. The volume of internal, east-west networking traffic in data centres is increasing.
  3. Data sets are increasing in size as people realise that, with AI and analysis tasks, the bigger the data set, the better the data. Big data needs to be sharded across hundreds of servers, which leads to even more east-west traffic.
  4. Security attacks are growing.

These issues are making data interchange between data centre nodes inefficient. For instance, an X86 server is good at application processing but not at data-centric processing, Sindhu argues. The latter is characterised by all of the work coming across network links, by IO dominating arithmetic and logic, and by multiple contexts which require a much higher rate of context-switching than typical applications.

Also, many analytics and columnar database workloads involve streaming data that need filtering for joins and map reduce sorting – data-centric computation, in other words. Neither GPUs nor X86 nor ARM CPUs can do this data-centric processing efficiently. Hence the need for dedicated DPU, Sindhu said. 

Fungible Video

DPUs can be pooled for use by CPUs and GPUs and should be programmable in a high-level language like C, according to Sindhu. They can also disaggregate server resources such as DRAM, Optane and SSDs and pool them for use in dynamically-composed servers. Fungible is thus both a network/storage/security DPU developer and a composable systems developer.

It’s as if Fungible architected a DPU to perform data-centric work and then realised that composable systems were instantiated and torn down using data-centric instructions too.

A Fungible white paper explains its ideas in more detail and a video provides an easy-to-digest intro.

We understand the company will launching its first products later this year and look forward to finding out more about their design, their performance and route to market.