This news has been a long time coming as Kioxia first announced its XL-Flash technology back in 2019. It used BiCS 4 flash with 96 layers, formatted as SLC (Single Level Cell; 1 bit/cell) and with a latency variously described as less than 25 and 20µs and down to 5µs.
Kioxia’s announcement contains no executive’s quote and says the FL6 bridges “the gap between DRAM and TLC-based drives, making them well-suited to latency-sensitive use cases such as caching layer, tiering and write logging.”
The FL6 has a PCIe Gen-4 interface and supports NVMe v1.4. We’re told it performs well on low queue depth workloads but its true strengths are revealed as workloads become more demanding and mixed. There it delivers a reliable quality of service.
Capacities are 800GB, 1.6TB or 3.2TB and the endurance is 60 drive writes per day — much better than TLC flash.
Kioxia provided a performance table;
For comparison Intel’s P5800X Optane SSD also supports PCIe Gen-4 and has a 100 DWPD rating. Its latency is under 6µs and its random read/write performance is 1.5 million IOPS for both reads and writes. The sequential performance is up to 7.2GB/sec reading and 6.2GB/sec writing, all of which makes it faster than the FL6 generally and much faster on random and sequential writes; 3.75x and 15.5x respectively.
A P5801X product goes faster still: 1.55 million random read IOPS, 1.6 million random write IOPS, up to 7.4GB/sec for sequential reads and writes.
You might cavil at Kioxia’s description of the drive as storage-class memory. Memory is accessed using load and store instructions, not through storage IO. Think of this in that case as a muscled-up dragster of an SSD.
The FL6 has dual ports for high-availability, a 2.5 million-hour MTBF rating and has SED and FIPOA 140-2 security options. We have not been told the warranty period but expect five years.
Socket-based licensing Veeam users are getting alarmed about Veeam Universal License Perpetual costs amidst Veeam’s plans to phase out socket-based licenses.
Certain Veeam data protection products, such as Veeam Backup & Recovery and the Veeam Availability Suite, could be licensed per socket and on a perpetual basis in the past. Veeam products could also be licensed using a Veeam Instance Licensing (VIL) arrangement. Such instance licensing was also used for Veeam public-cloud-based services. An instance is a unit or token of Veeam software workload capacity that is used in subscription pricing.
Veeam wants to phase out socket-based licensing. This is causing problems to some users. Poster “pirx”on the Veeam Community Forum complained of “a massive price increase (6 digit) due to the higher subscription renewal fee” and said there was a high frustration level about it.
When Veeam introduced its Veeam Universal License (VUL) this portable license could be used interchangeably across Veeam products and workloads, whether they be virtual, physical or in the cloud. They came in two flavors, both instance-based: Subscription and Perpetual. Existing socket-based Veeam licensees can face socket-to-instance conversion pricing problems when migrating to VUL, which have surfaced in the forum.
Just silly
Poster Chris Gundry said migrating from socket-based licensing to VUL Subscription was semi-reasonable in price and he might see a cost reduction over three years if he didn’t increase his count of backed-up virtual machines. But migrating to VUL Perpetual “was just silly in my opinion. Yr1 was 2.6x our current perpetual socket renewal cost and yr2 onwards renewal costs were 2.2x our current renewal costs, only marginally lower than the yr1 costs.”
Anton Gostev.
Anton Gostev, SVP for Product Management at Veeam, said there was a standard process for migrating Perpetual Socket licenses to VUL Subscription licenses but no similar process for migrating Perpetual Sockets to VUL Perpetual licenses.
Gostev posted: “Such migrations are currently done on case-by-case basis, as exceptions and via a custom process.” There was no standard process because there was insufficuent demand for one.
He said: “The biggest challenge they are facing is that no particular customer is a reflection of all customers as it comes to migrating off of a Socket license. For instance, with low density customers, migrating to VUL Subscription can be quite the savings and we see lots of such customers migrating. Also for lower edition customers, migrating to VUL can again be a great deal.”
For other customers this might not be the case. In that circumstance “Veeam should just let those remaining customers stay on Socket for as long as they want. With the caveat being periodic indexation of a Socket renewal price, to account for growing VM density per socket due to hardware advancements (like we’ve already been doing for some time).“
Pricing shock
But, because there’s no standard process users like “pirx” can react by looking to move away from Veeam when they get a price shock.
Gostev replied to “pirx” that “one of the core concepts of our VUL migration policy is a perpetual discount on future subscription renewals, which is specifically designed to ensure your subscription renewal costs are in line with what they were with the socket license.”
In fact: “If you like the VUL migration offer, take it. If you don’t like the offer, vote against it by staying on sockets — and so long as many customers keep refusing it, further migration policy changes will be made to make it more attractive.”
Unless Veeam can come up with acceptable case-by-case sockets-to-instance conversion pricing then Perpetual Sockets-based licensees will not move to VUL Perpetual instance-based licenses, and may even move away from Veeam altogether.
Apple’s iPhone 13, expected to be launched at the company’s California Streaming event on September 14, might have up to a terabyte of storage.
Wedbush analyst Daniel Ives said as much in a research note published last month, cited in MacRumours. He’d started forecasting the terabyte maximum storage capacity in March.
Analyst Ming-Chi Kuo concurred, saying the minimum capacity would be at 128GB — the same as the iPhone 12 Pro. From the iPhone 8 onwards and up to the iPhone 12, the minimum capacity has been 64GB.
It’s suspected that the phone’s uprated camera and higher-resolution video recording capability is behind the storage increase.
Minimum and maximum storage capacities for iPhone models.
The iPhone 12 continued the 256GB storage maximum started with the the iPhone 7 and carried on through the iPhone 8, 9, XR and then the 12 models. It was doubled to 512GB with the iPhone 12 Pro, which featured 10-bit HDR recording and 4K Dolby Vision HDR video at 24/30/60 fps.
In comparison the iPhone 12 was given 4K video at 24/30/60 fps, 1080p video at 30/60/120/240 fps, and Dolby Vision HDR (up to 30 fps). The iPhone 13 could get another increase in video recording resolution and size to cinematic video (subject in sharp focus, background blurred) and ProRes support (lossy video up to 8K resolution). An Apple white paper explains what this is in more detail.
It appears that, having killed the consumer still camera industry, the smartphone is going after video cameras. Low-end consumer camcorders have long been replaced by smartphones, but the rumoured specs for the iPhone well and truly rival the cameras used by professionals.
Wells Fargo analyst Aaron Rakers told subscribers smartphones account for 30–35 per cent of total NAND Flash bit consumption. He estimates: “Apple iPhone could account for up to 15 per cent of total NAND Flash bit demand.” That could make Apple the single largest flash industry customer in terms of capacity purchased.
DRAM is expensive, and servers have limited capacity through their socket architectures, but we need more DRAM as processors get more deeply involved in working on larger and larger datasets for applications like AI and analytics. High Bandwidth Memory (HBM) and four other approaches are being explored to solve the problem and increase DRAM’s areal density: cell flipping, modified NAND-cells, floating body cells, and gain cells.
They’re examined in a Semiconductor Engineering article entitled “Will Monolithic DRAM Happen?” We look at the cell flipping idea here. The other concepts seem less well advanced for now and we’ll set those aside.
We have looked at HBM in several articles. It involves stacking 2D Planar DRAM dies in layers of up to eight dies, interconnected with wire holes called Through Silicon Vias (TSVs) and connected via microbumps to an interposer chip which similarly connects to an adjacent processor. This architecture limits the number of stacks that can be hooked up to a processor.
High-Bandwidth Memory concept.
DRAM cell flipping
The 3D NAND concept involved changing the orientation of NAND cells to enable layers of them to be stacked inside a die.
3D NAND concept.
We can conceptually do a similar thing with DRAM — flipping the cells on their side. A DRAM cell has a small 2D area but a large, vertically oriented capacitor, making it high and difficult to stack in layers. As the 2D dimensions get smaller and the capacitor thinner, it has to be lengthened so as to hold sufficient charge. But if it is flipped onto its side and rotated 90 degrees, then the cells can be layered with a staircase design of bit lines per layer.
One approach to going vertical with the current DRAM cell. On the left is the standard layout of the cell, while the right shows it flipped onto its side. The capacitors are now horizontal, so enough layers are required to offset the increase in the lateral footprint. Source: Monolithic3D.
The photolithography patterning process used to make a layer during the DRAM manufacturing could be used for all the layers — so-called shared patterning — simplifying the foundry process.
3D DRAM using cell-flipping increases the 2D area of each cell, and a sufficiently high number of layers would be needed to increase the overall DRAM capacity for that area to make the exercise worthwhile.
More than one DRAM manufacturer is working on this cell-flipping concept, but none would comment directly about it.
Modified NAND-cells, floating body cells, and gain cell technology all involve abandoning the transistor-capacitor DRAM design, and so involve greater manufacturing changes and more difficult developments. Check them out by reading the Semiconductor Engineering article and following up its references.
Commvault is offering anti-ransomware services to its customers. NetApp wants a billion dollars a year in recurring revenue from the public cloud. Lenovo wants to offer all its products on-premises with a cloud-like as-a-service model. SingleStore has raised yet more cash for its do-it-all database — yet more evidence that VCs just love the data analytics field.
Commvault anti-ransomware services
Commvault has devised new anti-ransomware services. Its Ransomware Protection and Response Services provides resources and expertise to harden customers’ Commvault solutions from an attack, review the state of their data protection over time, and help them recover data when an attack does occur.
Ram Menashe, VP of Global Services at Commvault, said: “While we cannot guarantee our customers won’t be the target of an attack, we believe that by offering strategic protection plans, we can support customers at any stage of a ransomware attack — and do our best to help prevent it altogether with Commvault Protection Review.”
A Ransomware Protection Design and Plan service assists customers in validating their preparedness for ongoing threats and attacks.
A Ransomware Response Service provides the expertise and resources to help recover from an attack. As part of the service, Commvault provides a Ransomware Recovery Incident Manager backed by the Commvault Recovery Operations team. The service works with customers to identify and recover critical data and expedite a return to normal business operations.
These services complement Commvault’s Ransomware Protect and Recover products.
NetApp’s $1B ARR goal
With AWS FSx for ONTAP announced, NetApp’a EVP and GM of its Public Cloud Services business unit, Anthony Lye, briefed financial analysts. In his webcast presentation he said NetApp had a target of achieving public cloud services annual recurring revenue (ARR) of $450 million to $500 million in fiscal 2021 and $1 billion in fiscal 2025.
It is on track to getting to that level, and expects to make acquisitions as well as relying on organic growth to help drive its ARR upwards. Each sales person has a cloud business quota. As a reminder, NetApp is currently a $5.7 billion a year revenue company.
According to Wells Fargo analyst Aaron Rakers, Lye said that NetApp’s competition in this area came from object and block storage companies, and public cloud companies themselves.
Lenovo as-a-Service
$60 billion a year revenue PC to enterprise IT system supplier Lenovo has announced an everything-as-a-service strategy, under the TruScale brand name. This involves a scalable, cloud-like consumption model , fully managed services, and predictable payment options for hardware and service inclusions.
Lenovo is bringing all of its as-a-service offerings under one umbrella to provide a truly global solution that makes everything from the pocket to the cloud available via a single contract framework.
Research shows the as-a-service market is growing at four times the overall IT services total addressable market. In three years, as-a-service models will represent 12 per cent of enterprise x86 server spend and over 50 per cent of new enterprise storage spend, growing at 40 per cent CAGR and around 17 per cent of commercial PC spend — up from one per cent two years ago and growing at 50 per cent CAGR.
Lenovo has joined with infrastructure partners Deloitte, VMWare and Intel, and DaaS security partners Absolute Software and SentinelOne to provide organisations with various dynamic offerings such as hybrid cloud solutions for Edge to Cloud environments, best-in-class managed security solutions, and access to the most current storage infrastructure innovation, for the life of their data, with the ability to scale storage capacity on demand.
SingleStore fund raiser — again
SingleStore, selling a single database for all data-intensive applications, has raised $80 million in funding in a Series F round, taking total funding to $264 million. It raised $50 million in debt finance and $80 million in an E round in 2020. The company must be absolutely awash with cash.
Raj Verma.
Insight Partners led the round, with participation from new investor Hewlett-Packard Enterprise. Existing investors Khosla Ventures, Dell Capital, Rev IV, Glynn Capital, and Google Ventures also participated.
SingleStore said that a 300 per cent-plus increase in new customer acquisition for its cloud service and 150 per cent-plus year-over-year growth in cloud revenue led to raising a new round to continue to power innovation and growth.
SingleStore CEO Raj Verma said: “The status quo of stitching together multiple databases built for previous eras isn’t delivering the needed speed, simplicity, or efficiency. SingleStore is the industry’s first database built to solve the challenges of this new era in one database.”
Shorts
Acronis True Image has been rebranded as Cyber Protect Home Office, as it combines cyber security and data protection. Cyber Protect Home Office includes anti-malware proven to detect and stop the latest cyberthreats in real time, including zero-day attacks that have never been seen before. Protection covers videoconferencing applications like Zoom and Microsoft Teams, preventing attackers from accessing data in transit.
Box, under assault by activist investor Starboard Value, has won a board director election proxy war. Reuters reports shareholders rejected Starboard Value’s three candidates, re-electing Aaron Levie, Box’s co-founder and CEO, and Peter Leav and Dana Evan to the company’s ten-member board. This is a fairly rare setback for Starboard Value.
Open-source distributed NoSQL database supplier Couchbase has just released its latest earnings for its second quarter fiscal year ’22. Total revenue for the quarter was $29.7 million, an increase of 18 per cent year-over-year. Subscription revenue was $28 million, an increase of 19 per cent. Total ARR for the quarter was $115.2 million, an increase of 20 per cent. Gross margin for the quarter was 88.1 per cent.
Cribl, an Observability Pipeline startup, has raised $200 million in Series C funding led by Greylock and Redpoint Ventures. This brings Cribl’s total funding to $254 million, coming on the heels of sizable deals with large enterprise customers including FINRA, Rivian, and Cox Automotive. Cribl’s LogStream product parses and routes any type of event data to customers’ chosen analytics tools and storage destinations without fear of vendor lock-in, complementing tools such as Splunk, Datadog, and Exabeam.
DataOps supplier Delphix has gained SAP certification for the integration of its DevOps Data Platform 6.0 with SAP NetWeaver and SAP S/4HANA. Customers for these SAP products can use Delphix to automatically mask data for privacy compliance, secure data from ransomware attacks, and deliver efficient, virtualised data to accelerate SAP releases or train artificial intelligence and machine learning models.
Hitachi Vantara Australia & New Zealand today announced that it has successfully achieved a Data Centre Virtualisation VMware Master Services Competency. VMware Master Services Competencies are designed to help partners demonstrate customer-centric solutions and technical proficiency, with proven success and expertise in a specialised area of business.
IBM Spectrum Scale container native v5.1.1.3 is available now and features:
Automated deployment of CSI operator and driver
Support for storage cluster encryption
Support for rolling upgrade of IBM Spectrum Scale image versions
Support for a limited set of IBM Spectrum Scale configuration settings
Support for up to four storage clusters
Support for up to 16 remote filesystems
Automated IBM Spectrum Scale performance monitoring bridge for Grafana
Spectrum Scale CSI driver v2.3.0 is also available now:
Support for Red Hat OpenShift 4.8 and 1.21
Support for Kubernetes subpath
Migration of CRD to v1 from v1beta
Liveliness probe for CSI driver pods
Minimum Fileset based volume size is changed to 1Gi
Changes in inode calculation for preallocation
Images are now in IBM Cloud registry for CNSA based deployment
Automated CSI deployment with CNSA
Kingston Digital Europe is shipping a super-fast USB drive — the XS2000 — using USB 3.2 gen 2×2 speeds up to 2000MB/sec and capacities of 500GB, 1TB and 2TB. The 20000MB/sec speed needs a USB 3.2 gen 2×2 connector; it’s up to 1000MB/sec with a USB 3.2 gen 2 connector. This pocket-size drive connects to a PC, notebook or other portable device host via a USB Type-C connector. The XS2000 includes a removable ruggedised sleeve and IP55-rating to withstand water and dust. Expect prices around $99, $160 and $285 for the 500GB, 1TB and 2TB models.
Kingston XS2000.
LogDNA, which supplies a log management observability platform, unveiled early access for its coming LogDNA Streaming product. LogDNA’s Data Ingestion Pipeline can ingest, parse, and normalize massive, fluctuating amounts of structured and unstructured log data. LogDNA Streaming now automatically sends that data to any application or analysis tool, while maintaining tight control over storage costs and compliance measures.
A Nasuniblog by Tech Marketing VP Tom Rose dismisses AWS’s FSx for NetApp ONTAP, saying: “The difference between Nasuni and NetApp is the difference between a platform that was born in the cloud and one that has been ported to the cloud.” And he says: “The NetApp approach is to lift and shift the primary storage you’ve relied on in the past to the cloud, then add on costs for snapshots, cloud tiering, backup, DR, and edge caching.” Read the blog for more kick-the-competitor stuff.
Nebulon, an Infrastructure-as-a-Service vendor with a DPU-like server add-in card (Service Processing Units), has set up a Nebulon smartPartner programme with two new incentives, Nebulon smartStart and Nebulon smartRewards, designed to help reseller partners significantly accelerate and grow both their revenue and margins. Tim Pitcher, VP Sales at Nebulon, said: “It’s no secret that the channel industry has experienced limited margins and growth as a direct result of overly-distributed hyperconverged platforms.” So sell Nebulon kit instead.
PNY is offering branded HP USB 3.2 Flash Drives to its sales network: the HP x5600b & x5600c. These USB flash drives are designed to store, transfer and share digital files quickly and easily from a wide range of devices: smartphones, PCs, tablets, etc. They are available in capacities ranging from 32GB to 256GB. The HP x5600b USB flash drive has dual connectivity consisting of a standard USB type A connector and a micro-USB connector. The HP x5600c USB flash drive features dual connectivity consisting of a standard USB Type A connector and a fast USB Type C connector.
A Qumuloblog by Head of Competitive Intelligence David Chapa dismisses AWS’s FSx for NetApp ONTAP, saying: “AWS FSx is a managed service offered by AWS, and this makes the third such offering by AWS. Sounds like a big win, but it’s the same legacy complexity of on-prem but now on AWS.” Read the blog for more kick-the-competitor stuff.
Rumour central — it might just be rumour but we hear a certain SaaS backup mover and groover could execute an IPO manoeuvre in early 2022.
Backup and archive tape and disk storage vendor Spectra Logic announced LTO-9 tape availability in its tape library range, including the Spectra TFinity ExaScale, T950, T950v, T680, T380, T200, T120, T50e and Spectra Stack. Using LTO-9 cartridges the TFinity library can hold 1EB of uncompressed data — the first library in the world to do so.
Spectra T950 library now supports LTO-9 and its 45TB compressed data cartridges..
SvSAN storage software supplier StorMagic was included in the latest edition of the Gartner Voice of the Customer Report for Hyperconverged Infrastructure Software Vendors. Get a copy of the Gartner report here.
Archiving system supplier StrongBox Data Solutions uprated its channel program. The enhanced Ambassador Partner Program includes 100 per cent guaranteed deal registration and protection, easy-access partner portal, competitive margins and recurring revenue opportunities, marketing investment funds, sales support and qualified lead access.
Tuxera, a storage software and networking technology company, has agreed to acquire HCC Embedded, a developer of embedded file systems, flash management, and network data transfer software. The two say their combined experience and understanding of data storage and transfer technology covers any operating system or real-time operating system, any flash memory type, any hardware environment, and any storage interface, and a wide range of network protocols.
Flash chip, SSD and HDD supplier Western Digital announced that the Science Based Targets initiative (SBTi), has approved its greenhouse gas emissions reduction goals. SBTi has two different ambition levels: the Standard Commitment to limit global warming to well below 2°C above pre-industrial levels and a more demanding 1.5°C trajectory commitment. WD has committed to the 1.5°C path. It commits to reduce its Scope 1 and 2 emissions by 42 per cent by 2030, from a 2020 base year. The company is also adopting a Scope 3 target to reduce the emissions intensity of its products by 50 per cent by 2030.
Open-source distributed SQL database supplier Yugabyte has announced a new EMEA customer win: global digital money transfer service business WorldRemit.
IDC’s latest quarterly external storage market tracker shows a great year-on-year advance by Huawei, and a tremendous fall by IBM — but, really, nothing much has changed at all.
The worldwide enterprise external OEM storage market grew 9.7 per cent year-on-year to $6.93 billion, with Dell leading at $1.86 billion and 26.8 per cent market share. Joint third-ranked Huawei was the top growth performer at 26.7 per cent, its revenues rising from the year-ago $489 million to $619.5 million. Joint fifth-ranked IBM was the worst growth performer, with a 36 per cent drop from $504.6 million a year ago to $322.9 million.
Total external OEM storage capacity shipped was up 27.9 per cent year-over-year to 22.1 exabytes during the quarter. Research manager, Infrastructure Platforms and Technologies at IDC, Zsofia Madi-Szabo, provided an IDC statement about its tracker: “The external storage systems market recovery observed during first quarter of 2021 continued even stronger during the second quarter. The degree to which demand returned was highly influenced by the rate at which each regional economy has recovered from the difficulties associated with the global pandemic.”
An HPE statement said: “Recently, HPE unveiled HPE GreenLake offerings that represent HPE Storage’s transformation to a cloud-native data services business. We are already seeing tremendous customer uptake of HPE Alletra, a core component of the new offerings. We also continue to see strong uptake of HPE Nimble dHCI, which continues to disrupt the HCI market.”
So Alletra is off to a good start. That’s good news. HPE didn’t call out Alletra sales in its latest quarterly results, only mentioning Nimble and Primera.
Pure provided a statement as well: “According to the IDC data, in Q2, Pure executed on all fronts showing growth in every region. While the external OEM market showed a strong recovery in Q2, growing 9.7 per cent year-on-year, Pure did even better, growing 12 per cent. Pure also grew 12 per cent in the AFA market (the overall market grew 7.6 per cent).”
William Blair analyst Jason Ader said: “The market improvement was driven by a lessening impact on the economy from the pandemic and a release of pent-up demand for on-premises infrastructure, which was underpinned by strong second-quarter growth from storage leaders like NetApp, HPE, and Pure Storage.”
Here’s an IDC table showing its numbers for the top seven suppliers:
Going by this table, Huawei is doing fantastically well while IBM is heading downwards, ready to be overtaken by Pure Storage with its 12 per cent growth rate. The group of suppliers (rest of market) outside the top seven grew 17.6 per cent, so there must be smaller suppliers outgrowing most of the leaders.
Same old, same old
However, looking at the storage tracker trends from 2016 shows a different story, and we’ve charted it:
The thin lines are our inserted connecting lines where a supplier drops in and out of IDC’s tracker table.
Dell (+ EMC) is the clear and steady leader revenue-wise, though it has fallen behind the aggregate of the market since the second 2019 quarter. Then there is the following pack — a group of six suppliers, some of whom drop in and out of IDC’s tracking scheme. That’s because IDC actually tracks the top five suppliers but declares a statistical tie between suppliers whose revenues are within oneper cent of each other — hence the joint third positioning between NetApp, Hitachi and Huawei.
Our chart shows that Huawei’s storage revenues exhibit large rises and falls — it has overtaken Hitachi before, only to then fall back. Similarly with IBM. HPE, NetApp and Pure Storage are more consistent, with HPE and NetApp overtaking each other again and again. Pure Storage in fact has been close to IBM’s revenues before: Q1 2019, Q1 2020 and now Q2 2021. Each time, IBM has accelerated away, and may do so again, probably with help from the mainframe cycle and high-end, mainframe-attached array sales. It’s possible to discern a long-term trend for Pure Storage to catchup with IBM and potentially overtake it in the future — maybe after the next mainframe cycle.
However, the long-term trend has both Huawei and Pure Storage growing their revenues at a faster rate than the other pack suppliers. Pure Storage’s growth is more consistent than Huawei’s but it has not been kicked out of the US market — a shock with which Huawei has had to cope.
All-flash arrays accounted for $2.7 billion in revenues, up 7.6 per cent year-on-year. Hybrid flash/disk arrays exceeded this with 13.3 per cent growth to $2.8 billion.
Looking at the regions
Regional market growth rates varied widely:
China — up 33.3 per cent (benefitting Huawei we expect);
Japan — down 21.2 per cent (to the detriment of Hitachi);
Asia/Pacific (ex-China and Japan) — up 15.4 per cent;
Europe, Middle East and Africa — up 3.7 per cent;
Central and Eastern Europe — up 26 per cent;
USA — up 5.9 per cent;
Canada — up 29.5 per cent;
Latin America — up 12.5 per cent.
All in all the storage market grew. Huawei, HPE, Pure Storage, NetApp and Dell outgrew the market, while IBM and Hitachi did not. Let’s see what happens next quarter with a hopefully stronger recovery world-wide from the pandemic.
Fidelma Russo, VMware’s SVP and General Manager of its Cloud Services Business Unit, abruptly resigned from the company on 5th September, with some indications that she had also been appointed to the CTO role before leaving.
The news was confirmed by VMware, with a spokesperson telling us: “Fidelma has made a personal decision to leave VMware. We thank her for her contributions.”
Update 13 September: VMware did not deny, or confirm, that Fidelma Russo had been appointed CTO before she left.
Fidelma Russo.
In her SVP and GM roles she had joint responsibility with Mark Lohmeyer for VMware’s Cloud Services. Russo lead engineering, cloud operations, product management, product marketing for the VMware Cloud portfolio of services across Cloud Provider Partners, Local Cloud and the major Public Clouds. We had also heard that Russo was appointed as VMware’s Chief Technology Officer at the beginning of September and quit a few days later. We asked VMware about this yesterday and have been told: “As you are aware, Kit Colbert has been named our Chief Technology Officer.”
We then specifically asked VMware’s communications head whether Russo had been appointed CTO and left while in post, and received this reply: “I have no additional comment.” Make of that what you will.
Kit Colbert was appointed to the VMware CTO role on 8 September. On his appointment VMware CEO Rangarajan (Raghu) Raghuram said: “It is with great pleasure that we welcome Kit into the position of CTO at VMware. Kit’s passion, proven technological savvy and excellent leadership capabilities will be instrumental in advancing our innovation engine and research and development efforts.”
If Colbert was a second-string replacement for Fidelma Russo, that puts a different complexion on his appointment.
Greg Lavender, VMware’s previous CTO, was in the post from June 2019 until June 2021, leaving to join Intel as its corporate CTO and SVP/GM of its new Software and Advanced Technology Group, working for CEO Pat Gelsinger, VMware’s prior CEO. Gelsinger’s departure to Intel was followed by Raghu Raghuram’s ascension to the CEO spot.
Before being appointed CEO, Raghuram was COO for Products and Cloud Services at VMware, and would have worked with Russo.
Russo has had a stellar career:
Digital Equipment Corporation — 1984–1994, Design engineer to Senior Manager Alpha server;
Data General — 1995–1999, Group VP Aviion servers and software;
Dell EMC — 1999–2002, VP;
Sun Microsystems — 2003–2006, SVP and GM network storage and software;
HP — 2006–2007, VP Adaptive Infrastructure;
Sepaton — 2007–2010, COO;
Dell EMC — 2011–2017, SVP and GM enterprise storage and software;
Iron Mountain — 2017–2020, CTO and EVP global technology and operations;
VMware — 2020–2021, SVP and GM Cloud Services Business Unit.
What reason could there be for her to leave VMware so abruptly, days before a new CTO was appointed?
Dell EMC has worked with VMware and AWS to add EKS Anywhere to its hyperconverged VxRail systems, with storage needs fulfilled by either vSAN or PowerStore external arrays.
EKS (Elastic Kubernetes Service) is Amazon’s open-source Kubernetes distribution, enabling the management and deployment of stateless and stateful Kubernetes-orchestrated containers in the AWS cloud. EKS Anywhere is the software behind it and is available for use on-premises in vSphere virtual server environments. As VxRail systems run vSphere, they too can run EKS Anywhere.
Chad Dunn, Dell EMC VP of product management for hyperconverged and converged systems, blogs about this, saying: “Today’s announcement includes certification of two infrastructure configurations based on our fully curated and turnkey VxRail hyperconverged infrastructure (HCI) system.”
These are standard VxRail with vSAN virtual SAN software using storage drives in the HCI nodes and HCI with dynamic nodes where the storage resource is a Fibre Channel-connected block storage PowerStore array. The VxRail vSphere software runs Kubernetes clusters inside virtual machines and storage requests, routed across the Kubernetes Container Storage Interface (CSI) access virtual disks or vSAN file shares — both of which end up in VxRail’s physical storage. That means SSDs or disk drives in the HCI nodes or the shared access PowerStore array.
The equivalent VxRail EKS Anywhere PowerStore configuration.
Where could we go from here?
There is no direct link between EKS Anywhere and PowerStore and, therefore, no other Dell EMC storage arrays are supported.
However, we ask ourselves, since PowerStore can run VMware VMs inside the array, couldn’t the vSphere hypervisor involved in this run EKS Anywhere and — voilà — PowerStore is running EKS Anywhere directly. We checked with Dell EMC and a spokesperson said: “Hypothetically, yes, it could run it directly but it’s not available today and we don’t comment on specific roadmap plans.”
Dunn blogs: “Customers can run their Kubernetes orchestration in the public cloud or on-premises through a single console with EKS Anywhere, while at the same time having the reliability, security, ease of operations and global support that only Dell Technologies infrastructure can offer.”
Will Amazon support other hypervisors with EKS Anywhere? Hyper-V would pretty obviously be a “No,” but Red Hat’s KVM hypervisor and also Nutanix’s AHV might be candidates for such an extension. We might imagine AWS is receiving vigorous calls from these suppliers.
VMware could also work with other server OEMs to enable their vSphere-using HCI systems to support EKS Anywhere. That gets us thinking of HPE and its Nimble dHCI and SimpliVity HCI systems.
An AWS blog provides details on using EKS Anywhere.
ReRAM developer Weebit Nano has signed a deal with US-based Skywater to bring its technology into volume production.
Weebit’s ReRAM is a silicon-oxide material with electrical resistance set high or low by the presence or absence of silicon crystal filaments. The two resistance states signal binary one or zero. The resistance states are stable, making this ReRAM non-volatile, and state changes are faster than NAND writes, making it a storage-class memory tech. The technology has much longer endurance over write cycles than NAND as well, and needs less electricity to operate.
Coby Hanoch.
Weebit’s CEO Coby Hanoch supplied an announcement quote: “Our technology, which is highly efficient and extremely robust, is enabling greater performance and lower-power memory solutions for a range of new electronic devices. We look forward to collaborating with SkyWater — starting immediately — to ensure smooth technology transfer to their US production fab.”
Skywater is a so-called technology foundry for integrated circuit volume manufacturing. Its fab was originally set up by Minnesota-based Control Data Corporation in the 1980s, acquired by Cypress Semiconductor in 1991, and spun off into Skywater in 2017.
SkyWater and Weebit will jointly transfer Weebit’s ReRAM technology to SkyWater’s production fab and qualify it for volume manufacturing. Once that’s done, SkyWater intends to offer it to customers as embedded, non-volatile memory (NVM) IP on the company’s 130nm CMOS process.
Application areas include analog, power management, automotive, IoT and rad-hard designs. Weebit’s ReRAM is said to be cost effective, have enhanced endurance and retention at high temperature ranges, tolerant to radiation and electromagnetic fields, and cause zero interference with front-end-of-line (FEOL) analog components.
Skywater has licensed Weebit IP for use with customer designs.
If Weebit’s ReRAM becomes applicable for general use in servers then it could be the first real competitor to Intel’s Optane 3D XPoint technology to become available that is not a pseudo-NAND product — like Samsung’s Z-SSD. But we are a long way from this situation, and MRAM is being developed as well by Spin Memory and EverSpin. The EverSpin tech is available now and used as an SRAM replacement tech.
Dell has issued a data protection report saying customers are managing and protecting more data than ever, yet lack confidence in their data protection arrangements. In response it has added protection software and services including one that enables VMware VMs to be snapshotted without being stunned.
Dell’s 2021 Global Data Protection Index says business execs worry that existing data protection arrangements won’t meet future needs, and think that home working has increased data vulnerability. The survey also reveals — nice one this — the cost of data loss is 4x as high on average for organisations that work with multiple data protection vendors instead of one ($1.08 million compared to $287,000). Better to work with one then. Who might Dell’s report suggest? Hmmm …
Jeff Boudreau, President and General Manager, Infrastructure Solutions Group at Dell Technologies, offers a hint in a statement: “While ransomware attacks can be devastating for people and businesses, accepting defeat as a foregone conclusion is not the answer,” as Dell is the leading provider of data protection hardware and software.
Anyway, the two software advances both relate to the company’s PowerProtect appliances. Smart Scale is a management tool enabling up to 32 appliances to be managed as a single pool with up to three exabytes of logical capacity. Monitoring and managing a set of PowerProtect appliances is now easier.
However, Dell is not saying that PowerProtect appliances are clustered (scale-out) or that deduplication can work across the pooled appliances. But we have been told: “Deduplication is still limited to a single appliance. However, the storage unit mobility across appliances in a pool, combined with analytics, will allow us in the future to co-locate backup datasets that have better dedupe affinity on the same appliance, thus improving overall dedupe efficiencies in a pool.”
Stunning
Second, Transparent Snapshots have been added. These work on VMware virtual machines and are said to simplify and automate VM image-level backups, making them up to a claimed 5x faster and with up to 5x reduction in VM latency.
A spokesperson told us: “Previously, PowerProtect Data Manager used proxies in a given ESXi cluster to act as the Data Movers for the protection copies. A single proxy uses a minimum of four vCPUs and 8GB of RAM. Typically, multiple proxies will be deployed into a cluster based on the backup workload of the cluster (ie. the number of VMs that will need to be backed up, the size of the VMDKs, the duration of the backup window, etc.). Each of these proxies exists as a VM in the workload domain, thus those four vCPUs and 8GB of RAM, per proxy, comes from the resource pool that your other workloads share.”
So … now: “We use the Transparent Snapshots Data Mover (TSDM). TSDM is an extremely lightweight VMware signed VIB which is installed directly into each node in an ESXi cluster. Install footprint is a few megabytes, with a ‘max load’ footprint at 700MB. This releases a significant amount of workload domain resources back to the pool, which can now be used to run more of your application VMs. In addition, we made the TSDM management process transparent and automated. As your cluster scales up or down (in node count), PowerProtect Data Manager will automatically deploy new TSDM instances to new nodes, and clean up references to TSDM instances on removed nodes. Simple, clean, lightweight, and hassle free!”
And we were told: “The complete re-architecture of VM Image backups with Transparent Snapshots completely eliminates proxies of all types, replacing them with our lightweight TSDM module, embedded directly into each ESXi node, with autoscaling and zero-touch management.”
There are speed advantages to this approach. VM latency is the total round-trip time of a single disk I/O operation — the total time that an application needs to “wait” for a read or write operation to complete. With Transparent Snapshots, PowerProtect Data Manager 19.9 delivers up to 5x reduction in VM latency as compared to PowerProtect Data Manager with VADP. And there was a write latency reduction from 5ms response time with VADP to 0.97ms response time with Transparent Snapshots.
A Dell VMware slide deck from 2020 says transparent snapshotting takes place without disrupting VM execution (stunning) and making VMs unavailable.
A Dell video by Brian Reynolds, Senior Manager of Product Management, explains what’s going on:
Dell is also announcing a Managed Services for Cyber Recovery Solution in which Dell staff manage day-to-day cyber recovery vault operations and support recovery activities.
The 2021 Global Data Protection Index should be available here and a Dell blog provides background here.
PowerProtect Data Manager with Transparent Snapshots will be globally available this quarter, with no charge for customers with existing maintenance contracts. PowerProtect appliances with Smart Scale is in technology preview today and should be generally available in the first half of 2022. Dell Technologies’ Managed Services for Cyber Recovery Solutions are available globally today.
Open source Kubernetes-focused startup MayaData has undergone a CEO change as — this is almost laughable — the Chief Transformation Officer transforms the CEO who hired him out of the job and himself into it.
Evan Powell.
Coincidentally, this information comes out as Kubernetes startup Diamanta’s CEO change news broke. It’s all-change in Kubernetes startup CEO land.
Evan Powell, the ex-Nexenta CEO, became MayaData’s CEO and Chairman in June 2017 when the original ClouydByte startup renamed itself as MayaData. Powell built the company up with its OpenEBS product to the point where there was a deal with DataCore in February. Then DataCore technology was made available to MayaData and AME Cloud Ventures, DataCore Software, and venture capital and private equity firm Insight Partners invested $26 million in MayaData.
Insight Partners is also a DataCore investor. We might imagine some Insight Partners string-pulling is going on here — but we have no insight (ahem). Following the DataCore, deal Don Williams was hired as the COO and somewhat oddly named Chief Transformation Officer.
His LinkedIn profile says: “Joined to help MayaData scale in response to rapidly increasing adoption of OpenEBS for stateful workloads on Kubernetes. Responsibilities include insuring that every customer succeeds in their use of Kubernetes for data, with the assistance of OpenEBS and other MayaData software and services.”
Don Williams.
Fast forward four months and Powell is out, though still a board member, and Williams has the CEO reins in his hands. MayaData’s leadership webpage shows no sales, marketing or COO positions. Williams did what he was hired to do and, in effect, Powell hired his replacement.
This situation change is almost par for the course as far as Powell is concerned. He has a great early-stage startup CEO skillset, having built up Nexenta to the point where another CEO, Tarkan Maner, was appointed in 2013. That CEO then built it up more and helped sell the company to DDN in May 2019.
We might envisage Willians building up MayaData, on the foundations Powell laid, to the point where it gets acquired — possibly by DataCore through an Insight Partners push.
As we said yesterday commenting on Diamanti’s market situation, the Kubernetes storage world is intensely competitive, with suppliers such as NetApp Astra, Dell EMC, HPE Ezmeral, Pure-Portworx, not to mention SmartX, SUSE Rancher, StorageOS and many others duking it out for a piece of the action. It’s all good news for corporate recruitment agencies but means Williams has to get MayaData performing like a champ.
Executing a hyperconverged software to hyperconverged Kubernetes software pivot, startup Diamanti has changed its CEO — with a raft of consequent exec changes following the hiring.
Update: two more exec departures added. 10 Sep 2021. Hickey starting date changed to June 2021. 19 Jan 2022.
Prior incumbent Tom Barton assumed the CEO role in September 2018 after a less than stellar three-month episode as Tintri’s CEO in 2018, just as it was about to crash and get acquired by DDN. Barton quit Diamanti in May to co-found and run Astira in June. There is little information about this enterprise — not even a web site — and Barton’s LinkedIn profile reads: “Astira is changing the way satellites are designed, built, and managed.” Good luck with that.
Chris Hickey.
Diamanti’s board hired Chris Hickey to be the new CEO in June, the month after Barton resigned, implying no lengthy exec search process. This was an unexpected move as Hickey was previously the CEO of desktop publishing business Quark Software, being in that role for 28 months.
A person close to the situation said Hickey’s hiring was an interesting move by the board and many staff are fleeing. We note the following Diamanti exec changes:
CFO Jony Hartono left in September, replaced by Arnaldo Perez.
Communications Director Laura Finlayson resigned in August.
VP Product Brian Waldon left in June.
Chief Revenue Officer Andy Wild left in June, going to Mirantis.
VP Marketing Jenny Fong left in February to join Apptio in April.
COO Karthik Govindhasamy left in July to be the co-founder and CTO at Astira.
Founding Engineer Hiral Patel resigned in July 2021.
Field CTO/Product Evangelist Boris Kurktchiev went in August 2021.
San Jose-based Diamanti was founded in 2014 and has taken in $78 million in funding, with the last round being a $35 million C-round in 2019. That followed an $18 million B-round in 2017. Cash might be running tight. LInkedIn lists 92 employees. Its leadership web page shows no overall head of sales, no COO and no marketing head.
Taking a look at employee reviews on Glassdoor leaves one speechless, with “Cut throat snake pit” being one description and “Worst company in the world” being another. The embitterance factor on Glassdoor can be excessive as we all know, so let’s not pay overmuch attention to these eye-catching phrases, but still …
It looks as if the incoming CEO has a mountain to climb in terms of exec hires and sales coverage. The software technology looks good but Diamanti is facing competition from NetApp Astra, Dell EMC, HPE Ezmeral, Pure-Portworx, not to mention SmartX, SUSE Rancher, MayaData’s OpenEBS, StorageOS and many others. Diamanti’s employees need to pull together with a coherent and inspiring set of execs. Making that happen before competitors walk away with the market, is Hickey’s task. The eight exec leavers listed don’t think he can do it.
