Scale-out file system supplier Qumulo has beaten WEKA in the Azure cloud using an AI-related benchmark.
The SPECStorage Solution 2020 benchmark has an AI Image subset alongside four other workload scenarios: Electronic Design Automation (EDA), Genomics, Software Builds, and Video Data Acquisition (VDA). The Azure Native Qumulo (ANQ) offering achieved 704 AI Image jobs with an overall response time (ORT) of 0.84ms and delivering 68,849MB/sec
Qumulo claimed that this is both the industry’s fastest and most cost-effective cloud-native storage offering as its Azure run cost “~$400 list pricing” for a five-hour burst period. The software used a SaaS PAYG (pay as you go) model, in which metering stops when performance isn’t needed.
It said that deploying cost-effective AI training infrastructure in the public cloud requires transferring data from inexpensive and scalable object storage to limited and expensive file caches. ANQ acts as an intelligent caching data accelerator for the Azure object store, executing parallelized, pre-fetched reads, served directly from the Azure primitive infrastructure via the Qumulo file system to GPUs running AI training models.
WEKA recorded the highest results in four of the benchmark’s categories in January 2022, including AI. It reported 1,400 AI Image jobs with an overall response time (ORT) of 0.84ms using Samsung SSDs. A separate run with WEKA software running in the Azure public cloud recorded 700 AI Image jobs with a 0.85ms ORT and 68,318MB/sec. Qumulo has just beaten this by four jobs, 0.01ms, and 531MB/sec – a narrow margin.
A chart shows the differing vendor product results:
Qumulo has a lower ORT than WEKA in the AWS public cloud but a far lower AI Image job count.
A Qumulo blog claims that its Azure PAYG pricing model is disruptive. It argues that “most other vendors, including a previous submission at 700 jobs at 0.85ms ORT, do not communicate costs transparently.”
The blog authors further state: “They include a large, non-elastic deployment of over-sized VMs that you would have to keep running, even after deployment, in order to maintain your data set. They require a 1–3 year software subscription, costing hundreds of thousands of dollars, on a software entitlement vs having a PAYG consumption model.”
Tom “Pugs” Lyon, Sun’s eighth employee and NFS architect, argues that NFS must die and give way to a block-based protocol for large dataset sharing in the cloud.
He gave a talk at the 2024 Netherlands Local Unix User Group conference in Utrecht in February, which can be watched as a YouTube video starting from the 6:24:00 point.
Lyon first established his credentials with impressive stints at Sun, Ipsilon Networks, Netillion, Nuova Systems, and DriveScale:
His presentation’s starting point is that “file systems are great for files but they don’t really give a shit about datasets.” Datasets are collections of files. A dynamically mountable file system in Linux can be considered a dataset, and datasets, as used in AI training, can be made up from billions of files which are consumed (used) by thousands of GPUs, and also updated by a few hundred agents working on a few thousand files at a time.
Tom Lyon
NFS works and is popular, with its “killer feature that you can rapidly access arbitrary datasets of any size.” Its purpose is to provide a shared mutable data space. It provides access to large datasets, by network-unaware applications, without having to first copy them. But, Lyon argues, shared mutable data is basically a bad idea in concurrent and distributed programming. File sharing is not appropriate in the cloud.
It’s a bad idea because it’s error-prone, and developers are forced to use complex synchronization techniques to ensure data consistency, such as locks and semaphores. If we share immutable data instead, making copies of it, then there is no need for synchronizations, and data consistency can be guaranteed.
In the cloud, he says, sharing layered immutable data is the right way to do it, citing Git, Docker files, Delta Lake, Pachyderm, and LakeFS as successful examples. You can cache or replicate the data and it won’t change. But this can involve too much copying.
Admins should think about using NVMe over Fabrics and block storage providers to get over the copying issue. NVMe-oF provides “crazy fast remote block device access.” Block semantics are well defined and “aggressively cached by the OS.”
The POSIX-like, distributed file system BeyondFS “allows many different block storage providers” such as Amazon EBS, Azure, and GCP equivalents, OpenEBS, Dell EMC, NetApp, and Pure Storage.
There is much more in his presentation, which you can check out here.
Overall, Lyon proposes a new approach to achieve fast, highly scalable, and consistent access to dynamic datasets by using existing file systems, OverlayFS, and NVMe-oF. And he wants collaborators to join him in the effort of getting a new open source project started to achieve it. He can be reached at pugs@lyon-about.com or on Mastodon @aka_pugs@mastodon.social.
Bain Capital is to receive an $817 million investment payback from Nutanix, as well as almost 17 million shares worth $900 million.
Update: Rajiv Ramaswami quote counters Chris Evans’ view that the Bain payment may restrict Nutanix’ ability to invest in its business. 24 June 2024.
Nutanix went public back in 2016. In September 2020, it arranged a $750 million investment from Bain Capital Private Equity’s BCPW Nucleon unit via a 2.5 percent convertible senior notes deal due in 2026. These notes are a form of debt security that can be converted into cash and equity (shares) at a later date.
Dheeraj Pandey, Nutanix co-founder, chairman, and CEO at the time, said: “Bain Capital Private Equity has deep technology investing experience and a strong track record of helping companies scale. Bain Capital Private Equity’s investment represents a strong vote of confidence in our position as a leader in the hybrid cloud infrastructure (HCI) market and our profound culture of customer delight.”
Pandey retired shortly thereafter.
The deal is governed by an indenture which specifies that Nutanix is required to settle the conversion by paying $817.6 million in cash and delivering approximately 16.9 million shares of Class A common stock. Nutanix plans to use its existing balance sheet liquidity to settle the cash portion of the conversion and should deliver the shares in late July 2024 following regulatory approval.
When Nutanix took in the Bain investment, its revenues were $314 million to $328 million a quarter and it was making $185 million to $241 million losses per quarter. It finally became profitable in February this year and, as of its latest quarter end in June, had $1.651 billion in cash, cash equivalents, and short-term investments.
Rajiv Ramaswami
Nutanix CEO and president Rajiv Ramaswami said: “We appreciate the support, guidance, and counsel that Bain has provided us over the past several years and are pleased with their sustained endorsement.”
Analyst Chris Evans pointed out: “Effectively, for the issuing of $750 million, the company ‘pays back’ $1.73 billion (pays back in quotes as some of this is share issuance).
“It will be interesting to see the effect on the share price, but also on Nutanix’s ability to compete with the likes of VMware, when a big chunk of cash comes out of the company.
“Possibly the worst time for this to happen as it lets VMware off the hook somewhat. Nutanix might have to scale back sales and marketing spend, and maybe R&D, as these are currently the areas where the most money goes out the door. As reported in the third quarter 2024, the company only has about $1.6 billion in cash and short-term investments, so this payout will be quite a blow.”
However, Ramswami said: “Paying Bain the cash portion of the settlement of their recent note conversion does not mean we have to scale back investments in our business. We fund our investments in sales and marketing and R&D from the gross profit generated by our business, not from cash on our balance sheet. In the first nine months of our fiscal 2024, after covering all of our planned expenses, including sales & marketing and R&D, we generated over $370 million of free cash flow.”
Max de Groen and David Humphrey, partners at Bain Capital, will continue to serve as members of Nutanix’s Board of Directors. Humphrey said: “We have been really pleased with our partnership with Nutanix over the last several years, particularly during its transformation from a pioneer of hyperconverged infrastructure into a much broader hybrid multi-cloud platform provider. We continue to believe in the future of Nutanix. Their innovative technology, market position and operational discipline are enviable.”
De Groen said Bain had no plans to sell its Nutanix shares. The share price is up 3.63 percent to $54.01 over the past five days.
Rubrik grew revenues almost 40 percent in its first post-IPO quarter but made a $0.7 billion loss as it increased Opex spending 4.6x compared to a year ago, on R&D, sales and marketing, general and admin expenses.
Revenues for the cyber securing, protecting and recovery company in its first quarter of fiscal 2025, ended 30 April, were $187.3 million, up 38 percent annually, but with a loss of $723.1 million, contrasting markedly with the year-ago loss of $89.3 million. Subscription revenue rose 59 percent to $172.2 million and subscription annual recurring revenue was up 46 percent to $856.1 million. There were 1,859 customers with $100,000 or more in subscription ARR, up 41 percent year over year.
Bipul Sinha.
Rubrik co-founder and CEO Bipul Sinha said in the earnings call: “We got off to a very strong start to the year. It was an exceptional quarter.” And: “We benefited from some large transactions … We had two times more ARR from million dollar-plus transactions this quarter versus Q1 of fiscal 2024.”
His prepared statement on the results said: “Rubrik is a leader in one of the fastest growing new segments in the cybersecurity market, data security, with a total addressable market of about $53 billion by 2027…. While we are excited and proud of all the team has done to date, we are in the very early innings of data security and look forward to continuing to build an enduring company.”
Sinha said: “Our core assumption is that enterprises will get breached, we cannot stop this.” He reviewed the demand picture, saying: “As you look at the broader landscape, there has been a lot of investment in cyber prevention solutions around network endpoints, perimeters and cloud, but cyber recovery is little bit of an afterthought. Now folks are realising that recovery has been under invested and to ensure continuing operation they need to have solid cyber recovery because attacks are inevitable. So we are seeing significant strength in terms of the demand.”
In other words it is operating in “a strong market with high demand. … We’re still very early in a large market [with] 6,000-plus customers in a market size of 60,000 plus.”
Financial summary
Gross margin: 75%
Operating cash flow: -$31.4 million vs -$17.5 million a year ago
Cash, cash equivalents & short-term investments: $606.3 million
Debt: $297 million
The growth numbers are positive but the huge loss is surprising. A look at the numbers show that spending on three operating expense items rocketed compared to a year ago; research and development (R&D), sales and marketing, and general and administrative (G&A) expenses. Spending on the three items together totaled $816 million versus $184.5 million a year ago. That accounts for most of the loss.
B&F chart using Rubrik’s numbers.
Sinha commented on this as he answered a question in the earnings call: “When you look at the operating expense line items, starting with R&D; we are hiring …. And as we scale we are getting more leverage with the balance of great talent as well. And in sales and marketing … increasing … investments internationally and federal as well. We will see increasing leverage on the sales and marketing line. And lastly, we did step up G&A costs as a public company, but you will see benefits of scale as well.”
He’s growing the company’s capabilities because: “The explosion of data especially with AI is creating a massive market opportunity for Rubrik.”
We believe that Rubrik is focussing R&D work on AI, specifically on enabling Gen AI large language models to interrogate the masses of information in stored backups and also to help customers understand and improve their cyber-security arrangements.
Sinha talked about Rubrik’s Ruby chatbot: “the Gen AI agent for Ruby security cloud is focused on productivity for IT and security operations teams.” In contrast Cohesity’s Gaia chatbot is focussed on natural language, RAG-enhanced, access to information content in Cohesity’s stored backup data.
He said: “We are adding a lot of skill sets to the AI to ensure that that all the activities in this Rubrik security cloud can be done in natural language with ease of use, to be able to complete the task. It is as if we are shipping a person along with our product to handhold our customers to a tremendous outcome. It’s a long term project for us.”
We believe that Rubrik is also looking at RAG-enhanced, chatbot access to information in its data stores for customers.
Sinha also took a sideswipe at the Cohesity-Veritas acquisition: “I wish both parties on the merger luck, but the history is … there has not been a lot of success by merging a growth company with a legacy platform to create a new company.”
The outlook for Rubrik’s next quarter is $196 million in revenues, +/- $1 million, which will be a 29 percent increase year-on-year. The full fy2025 revenue outlook is $817 million +/- $7 million, a 30.1 percent increase annually at the mid-point.
With spiraling numbers of HPC and AI applications creating a shortage of GPUs in datacenters, a technical solution is needed to make sure organizations can maximize their GPU usage, while bringing more sustainability to GPU processing at the same time.
Update: This article has been edited to take out company sales figures, an estimated company valuation, and the potential value of a Series A funding round, all figures discussed at the original presentation, at Arc Compute’s request. 21 June 2024.
GPU hardware companies, and Nvidia in particular, have recently introduced very powerful platforms that place severe power demands on datacenters and on-premises deployments at large companies.
On an IT Press Tour of Silicon Valley this week to meet various data management companies, Blocks & Files encountered one startup that claims it has all the answers for the GPU availability and power conundrum.
Toronto-headquartered Arc Compute is a 17-person software company that is gearing up for a Series A round, on the back of being an early mover in the GPU sweet spot.
It is starting the process of patenting its software on a global basis, first starting in the US and then Canada and beyond.
Michael Buchel, CTO of Arc Compute, told us: “We specialize in the discovery of GPU inefficiencies, amid the rising dependency of high power compute and a GPU shortage. We want to achieve peak performance for customers and reduce the environmental impact.”
The go-to-market is direct for large strategic accounts, focusing on major AI/ML companies and organizations running supercomputers. That said, there is also a partner ecosystem that already includes many of the usual suspects in the high-compute field including Supermicro, Dell, HP, Nvidia, and AMD. Intel is so far not part of the party, said Buchel.
Arc Compute is also in discussions with major datacenter services providers to OEM its software bundle and, further down the line, plans to distribute the software through reseller channels in the datacenter space.
The pricing for its software is based per GPU, with a cost range of between $4,000 and $8,800 per GPU per year. There is also a public cloud provider cost of $0.37 per hour.
The Series A funding round may well include one or more of Arc Compute’s current customers, but Buchel would not comment on that. He confirmed, however, that he expects the round to be completed by the “end of this summer,” although it may well be delayed as a result of some big deals in the advanced pipeline. Arc Compute wants those deals to figure in its final valuation before the total of the funding round is finalized.
Buchel maintained the funding round was being done for “strategic reasons” and not out of any desperation for cash. “Some of the potential investors are very interested in this space and have large resources, so it would be good to get them on board.”
On the GPU sustainability front, Buchel said using the company’s utilization software could potentially lower the power usage of each GPU by around 30 percent in datacenters, once the ArcHPC Suite portfolio is fully developed.
It was clear from Buchel’s presentation that the company has already taken firm soundings from venture capitalists interested in a solution that could feature heavily in a very important niche in the GPU marketplace.
There are, of course, other technologies to help address the problem that Arc Compute is trying to solve. Addressing utilization with job schedulers and fractional GPU software are just two options. But Arc Compute outlined what it claims was the downside to these:
They cannot address low-level utilization points such as memory access latency where additional arithmetic operations could occur
They can lead to performance degradation
They cannot address fine-tuning of GPU environments for optimal task deployment for performance
They are unable to set or prioritize performance for “business objective alignment,” missing user governance policy settings for performance
The portfolio from Arc includes Nexus, Oracle, and Mercury. Nexus is a management solution for advanced GPU and other accelerated hardware. This software allows users to maximize user/task density and GPU performance. It achieves this by increasing throughputs to compute resources, while granularly tuning compute environments for task execution and providing recommendations for further improvements.
Oracle automates task matching and task deployment, and manages low-level operational execution of instructions in the HPC environment. It increases accelerated hardware performance through “scalable control.”
Mercury resolves task matching for the maximum number of unique tasks running. It selects hardware which will maximize the throughput for the average task running in the datacenter. It provides datacenter owners information to help scale their facilities to address new growing workloads.
NetApp has strengthened its sales team by hiring Mike Sakalas as the company’s vice president of US enterprise sales. Sakalas will report to NetApp’s Senior vice president of North America sales, Riccardo Di Blasio.
Sakalas will oversee the development and execution of NetApp’s enterprise business in the US, defining and implementing the company’s go-to-market strategy. Sakalas will partner closely with internal teams and NetApp partners to drive strategic customer engagements, said the vendor.
—
Data integrity firm Precisely has expanded its partnership with Databricks. Available now on Databricks Partner Connect, the Precisely Data Integrity Suite brings data management capabilities to Databricks environments that it claims address all aspects of data integrity, including data quality and observability, and geo addressing. The idea is that users can then leverage trustworthy data in their AI or advanced analytics initiatives.
“Precisely brings a unique combination of data integrity capabilities to our customer base,” said Roger Murff, VP of technology partners at Databricks. “Advanced data integrity tools from Precisely, combined with the Databricks Data Intelligence Platform, will provide our customers with faster access to reliable and actionable insights, driving innovation and operational efficiency.”
—
11:11 Systems, a managed infrastructure solutions provider, has announced the availability of 11:11 Cloud Object Storage for Amazon Web Services. The new offering will give 11:11’s customers the opportunity to cost-effectively store vast amounts of data with durability, scalability and security, it said.
“Storing mission-critical data in a single data centre creates the potential for a single point of failure,” said Justin Giardina, CTO, 11:11 Systems. “With cybersecurity threats looming larger than ever, using a storage solution with eleven 9s of data durability that redundantly stores data across multiple data centres is key.”
By default, 11:11 Cloud Object Storage for AWS redundantly stores a customer’s mission-critical data across a minimum of three data centres, as opposed to some other providers who always store data in a single physical building.
Customers will also be able to reserve capacity to secure storage discounts through a long-term contract, while still paying monthly. Additionally, the simplified and lower price point includes no fees for egress or API requests, and comes with 11:11 Systems’ 24/7/365 support.
—
Druva has announced enhancements across the Microsoft ecosystem with support for Microsoft Entra ID, which serves as the core identity and access management (IAM) for all Microsoft Cloud resources.
With support for Entra ID, Druva helps customers maintain business continuity in the event of cyber threats, data loss, corruption, or accidental deletion.
Formerly known as Azure Active Directory, Entra ID is a critical addition to Druva’s portfolio of supported workloads and reinforces Druva’s commitment to enhancing data security across the Microsoft cloud ecosystem.
Given its role as core IAM for all Microsoft Cloud resources, Entra ID is a prime target for cyber attacks. With Druva, customers can protect and quickly recover Microsoft Entra ID objects in the event of cyber threats, data loss, corruption, or accidental deletion.
“Druva has seen accelerating growth and increasing demand for Microsoft workloads, and we’re doubling down on our commitment to data security across Microsoft Azure environments,” said Nitin Nagpal, senior vice president and general manager, Druva Products.
Acronis has introduced Acronis XDR (extended detection and response), the newest addition to the company’s security solution portfolio.
“Easy to deploy, manage, and maintain,” said the provider, Acronis XDR expands on the current endpoint detection and response (EDR) offering, and delivers complete natively integrated, “highly efficient” cybersecurity with data protection, endpoint management, and automated recovery specifically built for managed service providers.
—
Virtuozzo has announced a new distribution partnership with Robson Communications, a cloud infrastructure and services specialist delivering solutions to managed service providers across the Americas.
Under the partnership, Robson now offers Virtuozzo’s “flexible, affordable alternative” to VMware cloud for MSPs and their customers, said Virtuozzo, enabling business continuity for companies impacted by changes to VMware’s bundling, licensing and partnership models, after its acquisition by Broadcom.
Robson provides cloud solutions tailored to each MSP’s technical and business requirements. The company will distribute Virtuozzo cloud solutions for on-premises deployment, for customers who operate their own datacenter infrastructure, and as on-demand infrastructure as-a-service and platform as-a-service hosted in Robson’s growing network of datacenters.
—
Continuing the VMware alternative theme, Red Hat has announced it has reduced the cost of its Virtualization Migration Assessment and Virtualization Training and Certification offerings, as it aims to help businesses navigate the “changing virtualization landscape” and “establish strategic paths” towards virtual machine (VM) infrastructure migrations. The services will be offered at a discounted rate through to August 31, 2024, and are designed to assist businesses with migration planning as they look to move off of “legacy virtualization providers”, and upskill IT teams to manage VMs using Red Hat technologies.
Red Hat’s Virtualization Migration Assessment is a two-week on-site, interactive discovery and design workshop with Red Hat experts, that will assess your business drivers, current state, and the “path to migration.” The assessment forms the basis for a high-level solution design, mapping out risks, building a path to production, and a customized migration journey and proposal.
Businesses that purchase Red Hat’s Virtualization Migration Assessment or an assessment from a qualified partner, and then move forward with subscribing to Red Hat Openshift Kubernetes Engine, Red Hat Advanced Cluster Management for Kubernetes and Red Hat Ansible Automation Platform, and services for the migration, are eligible to receive the total cost of the assessment credited back in the form of a subscription discount.
—
Oracle and Google Cloud have announced a partnership that gives customers the choice to combine Oracle Cloud Infrastructure (OCI) and Google Cloud technologies to help accelerate their application migrations and modernization.
Google Cloud’s Cross-Cloud Interconnect will be initially available for customer onboarding in 11 global regions, meaning customers to deploy general purpose workloads with no cross-cloud data transfer charges. Later this year, a new offering, Oracle Database@Google Cloud, will be available with the “highest level” of Oracle database and network performance, along with feature and pricing parity with OCI.
Both companies will jointly go to market with Oracle Database@Google Cloud, benefiting enterprises globally and across multiple industries, including financial services, healthcare, retail, and manufacturing, among others.
Ephemeral cloud environments such as serverless containers may be short-lived, but they present a real and growing security risk, and attackers can wreak havoc if they are unprotected.
Data security vendor SentinelOne has introduced its Singularity Cloud Workload Security for Serverless Containers, which it says provides real-time, AI-powered protection to secure containerized workloads running on AWS Fargate for Amazon ECS and Amazon EKS.
AWS Fargate is designed to let developers focus on building applications without managing servers and get ideas into production more quickly.
The sales pitch is that ephemeral containerized workloads running on AWS Fargate allow rapid scale and deployment to refresh environments, offering businesses technical agility. However, their short-lived nature does not automatically mean they are secure.
While these resources may only live for minutes, SentinelOne reckons attackers can compromise them within seconds and look for opportunities to move to higher-value, longer-living resources ahead of the ephemeral resource being deleted.
Adversaries can also gain an initial foothold elsewhere in a cloud environment and pivot to serverless container resources to conduct attacks.
Singularity Cloud Workload Security for Serverless Containers provides AI-powered runtime protection. It uses five autonomous detection engines to detect runtime threats such as ransomware, zero-days, and file-less exploits in real time, streamlining machine-speed response actions.
“As a strategic Amazon Partner Network member, we are committed to delivering market-leading innovations through simple integrations that enable customers to improve their security outcomes and change the game,” said Brian Lanigan, senior vice president, global ecosystem, SentinelOne.
Cloud Workload Security is part of SentinelOne’s cloud security portfolio, which includes Singularity Cloud Native Security and Singularity Cloud Data Security. The solution sits on top of the Singularity Platform and Singularity Data Lake.
The Singularity Platform is equipped with Purple AI, a generative AI security analyst that provides autonomous SecOps tools intended to accelerate security teams’ threat hunting and investigations, reduce Mean Time to Response, and deliver AI-powered enterprise security to stay ahead of attacks, says SentinelOne.
Cloud storage firm Hammerspace says it may well launch an initial public offering (IPO) in the “next 18 to 24 months” as it becomes “cashflow positive” this year, a prerequisite for any IPO.
Hammerspace delivers a global data environment which spans across data centres, and the AWS, Azure, and Google public clouds, providing unified file access to end users.
The technology is sold to both enterprises and service providers, with a broad 50/50 split in sales.
It was founded in 2018 by CEO David Flynn, and completed a Series A funding round in July 2023, raising a total of $56.7 million. Flynn co-founded flash hardware pioneer Fusion-io, which was eventually acquired by SanDisk in June 2014 before SanDisk was taken over by Western Digital in 2016.
David Flynn
During this week’s IT Press Tour of Silicon Valley, Hammerspace made new technology announcements and claimed strong sales growth. On the back of this, Blocks & Files asked Flynn about the potential of an IPO down the line.
He said: “We are making percentage margins in the mid-90s, and Hammerspace is a different business to Fusion-io, which was a hardware business, and hardware businesses cannot be scaled up as quick as software companies.
“We’re not a typical startup waiting to be acquired. We’ve done all the work ourselves and are building something important, and public ownership is important too.”
He added: “I had a phenomenal win with Fusion-io, which made me do some soul-searching, but I went back to technology, instead of lying on an island, or even buying one. This is not a vanity project, it is important to me.”
Before Fusion-io was acquired, it had its own successful IPO. On his time running a public company previously, Flynn said: “We only missed one quarter when I was leading it, and that was because Meta never opened a new datacenter that was planned, and which was going to use our hardware. We did fill that hole, but running a public company does leave some scars.”
That said, Flynn said Hammerspace’s high-margin software was pointing to a smoother future, whatever happens on the ownership front. He said since the firm had recently launched its hyperscale NAS (network attached storage) offering, the overall sales pipeline had mushroomed by a factor of ten, without having to hire any extra salespeople.
“We are building a very financially disciplined company, but we are reaching a point where we have to grow a lot faster to get in front of the opportunities being created by AI,” said Flynn. “It’s a tough market for an IPO right now, with budgets going down in some areas, and Nvidia sucking the air from everybody else at the moment.
“But, for the first half of this year, we have already generated sales that are ten times higher than what we made for the whole of last year. So, in 18 to 24 months, we could go for an IPO.”
This week, Hammerspace continued to add to its technology portfolio by launching advanced GPU data orchestration capabilities to accelerate access to S3 data. S3 applications can now connect to Hammerspace’s Global Data Platform, enabling object data to be automatically orchestrated to GPU resources, along with existing file data.
This expansion enhances accessibility to existing data sets in object storage and optimizes the pipeline across any storage type, “allowing infrastructure teams to focus on deriving insights and driving innovation.”
Molly Presley, SVP of global marketing at Hammerspace, said: “Accessing available GPUs in an organization’s own datacenters or in the cloud is a challenge. Even more difficult can be identifying useful data sets and placing that data local to the available compute resources.
“HPC centers and enterprise infrastructure architects are urgently looking for solutions to organize large data sets and mobilize them to where the GPUs are located. With the addition of the S3 interface, they can now quickly do it.”
Index Engines is an AI-powered analytics engine designed to detect data corruption and ransomware threats. Its CyberSense product is currently deployed at thousands of organizations worldwide, sold through strategic partnerships.
On this week’s IT Press Tour across Silicon Valley, the company said it intends to significantly scale up its channel partner reach, building on its established technology partner relationships with Dell, IBM, and Infinidat.
Dell sells CyberSense along with its PowerProtect Cyber Recovery product, and Cybersense is white-labeled alongside Infinidat Infinisafe with Cyber Detection and IBM Storage Sentinel.
After 20 years in the market, Index Engines has an installed base of 1,400 large organizations. After initially targeting storage companies to reach end customers through strategic partnerships, it is now considering system integrators and managed service providers to help win more business.
That said, five additional partners are imminently expected to join the existing big three, and they include other storage companies.
Jim McGann, VP of strategic partnerships at Index Engines, told the IT Press Tour: “With the increased ransomware threat, managing cyber liability needs to be a priority, and backup is not enough. While you have to have the ability to recover, data integrity is key, and you need to know that your data is reliable, which is what we address with CyberSense.”
McGann said data integrity requires content analysis, involving the continuous inspection of files and databases, the search for data corruption patterns indicative of ransomware, and the utilization of hundreds of data points with AI-based machine learning.
This current quarter, CyberSense was upgraded to v8.6, with an updated alerts page, including all infections found and new threshold alerts. There is also a dashboard to analyze current and previous alerts. In addition, a new hosts page allows users to view status of selected hosts and it indicates daily activity with intuitive graphs.
We were also given a sneak preview of further substantial improvements in the forthcoming v8.7 due later this year, but we can’t make them public at the moment. Future releases will aid the channel expansion, however. Also, there will be an announcement from Index Engines on June 18 designed to further ratchet up its market reach.
As the channel expansion comes to fruition, McGann said the current pricing model would continue, priced per TB of data analyzed, with CyberSense sold over one, three, and five-year terms.
McGann said: “Our model is working, we are privately financed with no debt, and we use everything we have developed ourselves over the last 20 years. Dell and IBM tell us there is no way they could have developed what we have.”
As the top three server OEMs – Dell, Lenovo, and HPE – announced extended contracts with VMware to continue selling hyperconverged systems and servers with the VMware stack pre-installed, plucky edge object storage vendor StorMagic launched its alternative and “cheaper” HCI solution for smaller data footprints.
On the OEMs’ VMware deal, StorMagic chief product officer Bruce Kornfeld said: “Their new agreement is another data point proving that HCI (hyperconverged infrastructure) is a highly desired way for end users to deploy IT solutions. The OEMs were desperate to be able to sell VMware because they don’t have hypervisors of their own.”
Bruce Kornfeld
Even with these new agreements in place, said Kornfeld, VMware remains “too expensive” for edge and SMB use cases. “Our purpose-built and cost-effective solution for edge deployments and branch offices at large organizations isn’t dependent on VMware.”
Broadcom’s VMware license strategy suggests that only the biggest customers are welcome: it made the operation of 3,500 cores running VMware Foundation Cloud the base requisite for enrolment in the partner programe, effectively shutting out small enterprise customers, resellers and CSPs.
Kornfeld said his company, like other VMware rivals, was looking to take advantage of customer and channel uncertainty around big changes to VMware pricing and contracts following Broadcom’s acquisition of the virtualization market leader, which he described as a “fiasco.”
StorMagic this week unveiled SvHCI, combining a hypervisor and virtual networking with the provider’s SvSAN virtual storage software, which is already used by thousands of customers around the world. The full-stack HCI also includes StorMagic’s in-house 24x7x365 customer support.
According to the recently published “Gartner Market Guide for Full-Stack Hyperconverged Infrastructure Software” report, 30 percent of the non-market leading full-stack HCI installed base in 2024 is projected to increase to 60 percent by 2029.
SvHCI is a software solution that installs directly on new or existing servers. It is designed to simplify operations and deliver high availability for edge and SMB environments with only two servers, while claiming to lower software costs by “up to 62 percent over VMware.”
Generally available later this summer, SvHCI is priced starting at $2,049 for a one-year subscription for a single server and up to 2 TB of storage. StorMagic has launched an SvHCI Global Beta Program to allow customers and partners to get early access to the software and test it on their own servers.
The range of savings StorMagic talks about is 21 percent to 62 percent. Comparisons include SvHCI one, three, and five-year subscription lengths, plus maximum storage capacities of 2 TB, 6 TB, 12 TB, 24 TB, and 48 TB. VMware comparisons are for the same subscription length and storage capacity of VMware vSphere Foundation with VMware vSAN and VMware Cloud Foundation with VMware vSAN.
Data management and observability vendor Apica has upgraded its Ascent Platform following its recent acquisitions of data fabric firm LOGIQ.AI and telemetry data management specialist Circonus.
Apica has integrated capabilities from both organizations into its platform to provide “deeper insights” into data management, it says. The aim is to simplify how customers gather data and “seamlessly” integrate it into their systems and platforms.
In addition to making two acquisitions over the last ten months, Apica announced $16 million in new capital funding during the period. The extra cash was used to expand the firm’s Office of the CTO, as well as to strengthen its engineering and research and development teams. These actions and the integrations have led to the launch of Apica Ascent 2.0.
“Organizations struggle to manage tool sprawl, so we’ve provided a platform that ensures their legacy and digital transformation strategies work well together,” said Ranjan Parthasarathy, chief technology and product officer at Apica. “Firms don’t have to leave their legacy tools behind while modernizing, Apica can help make everything work together to provide insights needed for the business.”
Parthasarathy maintained that Ascent was “easy” to implement, use, and manage, and that its cost was “low”.
The platform encompasses a wide range of features including network and overall performance, compute metrics, memory and storage, and file systems, among other areas. It gathers and uses application data like logs and metrics, and traces from organizations’ runtime applications.
“We support an approach that does not stick our clients in a walled garden,” added Parthasarathy. “With the onset of generative AI, we now face even more data growth, and customers will need open-source solutions like ours to keep costs down, and avoid a situation where data is closed off.”
Apica says it embraces “anything and everything” to support ecosystem connectors, enabling open protocols and data collectors.
The company is headquartered in Stockholm, Sweden, and El Segundo in California. It also has offices in London and Bengaluru, India.
There’s a new data transformation kid on the block in the form of Tobiko Data, which has just raised $21.8 million in funding.
Blocks & Files met Tobiko Data co-founder and CTO Toby Mao this week in Santa Clara, California, to discuss the firm’s position in the market.
Tobiko, named for the flying fish roe popularly used in sushi, develops SQLMesh, an open source data transformation platform. With SQLMesh, data scientists and analysts can build “correct and efficient data pipelines,” promises the firm.
Development environments can take too long to spin up and are costly, says Tobiko, and when things go wrong, it’s “painful” to undo changes. Also, there is a lack of visibility into data pipeline performance, and managing large datasets have their own complexities.
Fundamentally, says Tobiko, DBT projects “don’t scale up”. DBT (data build tool) automatically generates documentation around descriptions, models dependencies, model SQL, sources, and tests. DBT creates lineage graphs of the data pipeline, and aims to provide transparency and visibility into what the data is describing, how it was produced, as well as how it maps to business logic, but some tools are better than others.
Tobiko claims it can save customers time and money in the whole process by only having to build tables once, reducing warehouse costs, and helping to get more work done.
Formed around 18 months ago, with the founders having helped lead the development of data performance metrics at the likes of Apple, Netflix, Airbnb, and Google, Tobiko competes against more established data transformation companies like DBT Labs.
On this week’s IT Press Tour across Silicon Valley, Mao told press and analysts: “With Snowflake [the cloud data platform], if you use it inefficiently, it can cost you a lot of money. With our SQLMesh, you only have to produce tables once, while other technologies make you do them time and again.
“A single change in a query can affect billions of rows of data, and companies spend millions every year on unnecessary rebuilds of the warehouse when only a small precise change is needed.”
SQLMesh can be used for free, but the firm has just launched SQLMesh Enterprise, a paid-for product which is a full observability platform. It not only tells users something went wrong with their data, but it also tells users why.
Mao says: “We are talking to Snowflake, for instance, about the value of our technology, but the software is not currently integrated into their ecosystem. But it can save users thousands of dollars off their Snowflake bill.”
He added: “Your developers can do things quicker and be more productive. They don’t have to go and get a coffee every time they change something in the data, they just run things with us and carry on with what they’re doing, they don’t have to wait around.
“This is all down to our SQL expertise, including our own SQLGlot framework that supports SQLMesh.” SQLGlot has been made open source too. It is a SQL parser, transpiler, and translator that currently supports 24 different SQL dialects.
Tobiko’s advantage is its fundamental semantic understanding of SQL, meaning it only executes necessary downstream changes instead of completely rebuilding the warehouse. It also holds state, allowing for first-class incremental refreshes, because it knows where users left off, and powers virtual data environments. The system understands and remembers every version of every model, avoiding duplicative computation.
Automated data movement provider Fivetran is a fan and user of Tobiko technology, with founder George Fraser being part of the latest funding round.
Mao says Tobiko will remain focused on data transformation. “We are a dev tool, we are not a data warehouse, we do metadata and support the tooling to move the data between any platform, whether it be Snowflake, Databricks or anywhere else, without being locked in.”
On the potential of being acquired by a larger data management player, Mao said: “At the moment we’re having fun, I’m doing software with my friends. We’ll see where we are in the future.”
