Home Blog Page 340

NetApp posts strong Q1, plots big re-organisation

By
Chris Mellor
-

NetApp has pulled off a strong first quarter, posting revenues up five per cent y/y to $1.3bn and net income of $77m.

CEO George Kurian said yesterday: “In the face of an uncertain environment, NetApp performed well in the quarter with revenue, operating margin and EPS all exceeding our guidance… The improvements we made to sales coverage in FY ’20 and the tighter focus on execution are starting to pay off.”

We’ll return to the numbers later, but let’s take a look first at the NetApp re-org, announced yesterday.

NetApp has confirmed the company is laying off 5.5 per cent of the workforce – approximately 700 employees (see our report: NetApp layoffs hit SolidFire hard).

The company told us via an email statement: “In the realignment of our resources at NetApp, we are changing our approach to advanced technology and our industry standards body participation.We’re tightly aligning those functions to the business so that we can be more responsive to the rapidly changing demands of the market and customers.

“As a result, many Advanced Technology Group (ATG) and Standards and Industry Associations Group (SIAG) team members will be integrated into the business and engineering teams, while some will be leaving NetApp.”

NetApp’s Solidfire operation is shrinking to focus on core enterprise usage. “We are narrowing our focus with a SolidFire and HCI portfolio to the high-margin parts of the market,” Kurian said in yesterday’s earnings call.

Solidfire AFAs will get investment to increase operational simplicity and security plus additional software features. NetApp told us: “We are investing in NetApp HCI as a simplified and automated infrastructure solution for on-premises Kubernetes environments and are closely aligning NetApp HCI to Project Astra.”

The company is winding down efforts on MAX Data technology, which links host server Optane drives to ONTAP arrays – at least for the present. NetApp said: “Our launch of Memory Accelerated Data (MAX Data) was far ahead of the market. To give the market time to catch up with our innovation, we are focusing our resources to explore new ways we can use the core technology and underlying IP across our core storage and cloud portfolio.“

The numbers

OK, so what’s new?

NetApp has two strategic initiatives: to regain growth in its storage business and to push ahead with public cloud services. Kurian said in the earnings call: “Enterprises are prioritising transformational and hybrid cloud projects which drove our momentum as customers turn to NetApp to help them accelerate these plans.”

The Q1 results show flash storage sales are picking up, and cloud is promising – albeit from a small base compared with some competitors. In the earnings call yesterday, Berry said: “Public cloud services delivered an impressive $178m in ARR (Annual Recurring Revenue), growing 192 per cent year over year.” Three recent acquisitions, Spot, CloudJumper and Talon, accounted for $44m of the revenue growth.

Product revenues were $627m, down three per cent y/y. Hardware revenues were $316m, down seven per cent due to lower disk sales. Software product revenues were $311m, up two per cent as all-flash ONTAP revenues rose.

Berry said: “Our all-flash revenue of $567m was up 34 per cent year over year, nicely positioning us for share gains in the quarter.” The percentage of the installed base that has converted to all-flash was 22 per cent in Q1’19, 24 per cent in Q4’20 and 25 per cent in the latest quarter.

Kurian said in the earnings call: “Our installed base continues to grow. And the fact that with strong growth rates in all-flash array, we move the penetration one per cent at a time is just the dimension of the magnitude of our growing installed base. These are very, very big installed base numbers … One flash array replaces a few disk-based systems.”

Maintenance revenues climbed 14 per cent to $608m, accounting for 47 per cent of total revenue in the quarter. Berry said: “When combined, software revenue and recurring maintenance revenue totalled $919m in Q1, representing 71 per cent of total revenue.” The gross margin for these two areas is 83.4 per cent.

Revenue outlook for next quarter is $1.225bn – $1.375bn which at the midpoint implies a five per cent decline. Kurian said: “Enterprises are prioritising transformational and hybrid cloud projects which drove our momentum as customers turn to NetApp to help them accelerate these plans. … Uncertainty remains high, but we are moving into a new normal and adjusting to operate in a virtual environment.”

The company has plenty of wiggle room. Cash, cash equivalents and investments were $3.77bn at the end of the quarter.

Cohesity speeds up flash performance, hangs out with more hardware vendors

By
Chris Mellor
-
High speed business and technology concept, Acceleration super fast speedy motion blur of train station for background design.

Cohesity has accelerated its software with ‘I/O Boost’, a flash technology that, it says, improves backup and access performance by two-to-eight times compared with disk drives.

Matt Waxman.

Matt Waxman, VP of product management, said: “The Cohesity DataPlatform with I/O Boost makes flash even more affordable, while meeting increasingly demanding SLAs for backup, recovery, file and object services, and data insights.”

I/O Boost intelligently places data and metadata on a server’s directly-connected NVMe SSDs, the company says, striping data across physical media and enhancing the durability and cost-effectiveness of the SSDs through global wear-levelling. It also enables parallel access to metadata and data for faster overall storage IO. 

This accelerates metadata-intensive operations such as dedupe, replication, backup, etc. for faster data protection and disaster recovery. The technology optimises QoS policies, thresholds, and dynamically throttles data management tasks for predictable application performance across workloads.

Playing with the Big Boys now

Cohesity, a secondary data hyperconverger, has teamed with Cisco and HPE to enable some of their servers to the flash-optimised Cohesity Data Platform (CDP) software. The Cisco UCS C220 M5 and HPE ProLiant DL360 G10 are 1RU systems with 10 slots for NVMe drives. HPE is OEMing the Cohesity software while Cisco has added it to its general price list. Cisco and HPE channel partners can each sell the combined products.

Cisco said its HyperFlex hyperconverged software can run on the same server as Cohesity CDP, providing a happy convergence of hyperconverged appliance and hyperconverged data management software.

Cohesity has also certified Dell PowerEdge R640 and Intel R1208WF 1U x 8 NVMe drive slots to run the updated CDP software.

These hardware partnerships follow closely behind the Pure-Cohesity FlashRecover tie-in, announced earlier this month.

.

VAST Data boasts of big sales momentum

By
Chris Mellor
-
Tom Mendoza

VAST Data, the lavishly funded all-flash array startup, has announced Q2 revenue growth of 490 per cent, without revealing numbers.

Renen Hallak, VAST Data CEO, said today: “The adoption of Universal Storage continues to accelerate globally. In fact, we were above our very aggressive revenue target for both Q1 and Q2 of this year.”

In April, VAST said its first year of sales were were significantly higher than any storage vendor in IT history without revealing numbers. This sales momentum was good enough to galvanise a $100m VC funding round valuing the company at $1.2bn. 

The second 2020 quarter has continued that momentum. VAST says the business surge was due to several large orders from new customers and continued spend from existing customers. Two-thirds of existing customers bought more VAST storage, tripling their original purchases on average. It doubled its number of channel partners Y/Y, and has 100 open job positions.

VAST Data aims to consolidates all enterprise storage tiers onto its single scale-out platform. the company supplies a quad-level flash Universal Storage array, using NVMe-over-Fabrics, Optane SSDs and metadata management and data reduction features. The company claims low latency, high performance throughput at a cost equivalent to the price of enterprise disk storage.

VAST is representative of the rise on the popularity of QLC flash, as vendors extend endurance to enterprise levels through better data write management and wear equalisation across NAND dies in the drives.

For example:

  • Pure Storage this week announced a quadrupling of capacity in its QLC-based FlashArray//C with 24TB drives and coming 49TB ones.
  • Nimbus Data today announced its huge 64TB capacity ExaDrive NL using QLC flash today.
  • StorONE announced a QLC flash array with Optane SSD acceleration in June.

Mendoza

Tom Mendoza

VAST Data has recruited Tom Mendoza, ex-NetApp president and vice chairman, to the board. Mendoza joined NetApp in 1994 and ran sales for six years, becoming President in 2000 and vice-chairman 2019. He retired from the company in 2019 and now sits on several boards.

Mendoza provided an announcement quote: “New applications are driving a set of requirements in data infrastructure that we have never before seen. The market is recognising this as it moves beyond legacy storage offerings and VAST is uniquely positioned to lead this transition. I am thrilled to join VAST’s board and look forward to building the next great independent infrastructure company together.”

Nimbus Data touts 64TB SSD as nearline drive alternative

By
Chris Mellor
-

Nimbus Data today announced a 64TB SSD using QLC (4bits/cell flash, the highest-capacity QLC SSD available).

Nimbus is pitching the ExaDrive NL as a disk drive replacement and claims customers can increase storage capacity by 4x, improve data access times by over 100x, and reduce power per terabyte by 75 per cent compared to disk. The drive hits a sweet spot between nearline disk drives and NVMe SSDs, according to the company – it is faster than disk drives and slower than NVMe SSDs but has higher capacity than both.

Nimbus Data CEO Thomas Isakovich told Blocks & Files: “ExaDrive NL delivers the capacity and value to effectively replace hard drives in environments where performance is critical but where SSDs have historically been cost-prohibitive.”

The ExaDrive NL comes is 16TB, 32TB and 64TB capacity points and retail prices at time of writing are $2,900, $5,600 and $10,900. Nimbus updates prices every three months, based on flash market conditions.

Jeff Janukowicz, an IDC research VP, provided a quote: “High capacity QLC flash SSDs, like Nimbus Data’s ExaDrive NL, help organisations migrate to enterprise-grade flash storage as cost-effectively as possible, while simplifying hybrid storage that blends HDDs and SSDs seamlessly to optimise cost and performance for their workloads.” 

SSDs are cheaper to operate than disk drives, needing less power and cooling, and are much faster to access. SSDs are also more reliable. But they cost a lot more – the price differential today between enterprise SSDs and nearline hard drives is typically about 10x. Wells Fargo senior analyst Aaron Rakers argues customers will mass-replace nearline drives with SSDs when prices for the latter drop to a maximum of 5x higher than HDDs.

Isakovich tells us: “We are seeing $27/TB in distribution for the new WD 18TB nearline HDD. The 64TB ExaDrive NL series SSD … is priced at $170/TB in distribution. That is a 6x premium. So looks like we are right at the inflection point. Most analysts expect QLC pricing to decline in the coming quarters, so I believe a 5x premium will be achievable by year-end.”

The ExaDrive NL has a 3.5-inch case and gets its 64TB of data out to accessing hosts via SATA or a dual-port SAS interface, which supports high-availability enterprise needs.

Performance numbers are:

  • 100,000 IOPS (read and write @ 4 KB)
  • 500 MB/sec (read and write)
  • 0.2- 0.6 Drive Writes per Day (depending on block-size, up to 70 PBW, which would be the 64 TB at 0.6 DWPD)

The ExaDrive NL is qualified with servers from Dell EMC, HPE, Lenovo, Cisco and Supermicro and is available now from Nimbus distributors and partners.

HPE storage revenues recover but are still down

By
Chris Mellor
-

HPE’s third fy21 quarter storage revenues of $1.13bn recovered from the Covid-19 beating they took in Q2 but are still down 10 per cent year-on-year.

A huge order backlog was partly to blame for the company’s woeful performance in Q2. HPE cleared a big proportion of the backlog in Q3, which helped the company record a 4 per cent quarter on quarter revenue rise, from $1.086bn to $1.28bn. The results show it is not out of the pit yet but at least it appears to have stopped digging.

HPE storage revenues have declined for four quarters in a row on a Y/Y basis, as this chart shows: 

They have been trending down for some time before but HPE changed its financial reporting metrics, which rendered our historical numbers invalid.

There were several storage bright spots. The company saw a 31 per cent Y/Y rise in big data storage, where its product is built on MapR technology and Apollo servers. HPE reported 104 new customers for its recently refreshed Primera array, the successor to 3PAR, with a 114 per cent revenue rise. Encouragingly, a quarter of the new Primera customers were new to HPE. There was also 112 per cent growth in Nimble dHCI revenues year on year. HPE did not call out 3PAR, Nimble or SimpliVity product revenue.

The composable cloud or Synergy business revenues grew 30 per cent Q/Q, but that revenue is included in HPE’s compute segmente.

In the earnings call, CEO Antonio Neri identified that storage is often tied to compute for HPE. “A lot of our storage platform is basically a compute platform with software, right. Big data storage is a great example of that. It is an Apollo platform that runs our [MapR] software on it.“

He added: “We continue to strengthen our core capabilities in storage and compute, which are essential resources to store and process customers’ data. Every 60 seconds, we ship 46 terabytes of storage in four servers.”

HPE is transitioning to a subscription-based, everything-as-a-service business with its GreenLake offering. Neri said: “GreenLake services orders grew a record 82 per cent year-over-year. We believe this is faster than the orders growth of public cloud vendors, … While others are now publicly declaring plans to offer everything-as-a-service, we have been focused on this for several years and have made significant organic and inorganic investments to deliver a differentiated experience for our customers.”

Growing storage revenues while moving to a subscription business and away from a perpetual license model, which has more revenue recognised at the time of sale, will be challenging to HPE. The company also needs to raise its storage array game to better compete with the likes of Pure Storage.

Pure Storage revenue growth slams into Covid-19 wall

By
Chris Mellor
-

Covid-19 pandemic issues substantially curtailed Pure’s revenue growth in the second 2021 quarter.

Revenues grew two per cent to $403.7m Y/Y and losses improved slightly from $66m to $64.98m. This is much lower growth than the double figure percentages Pure has consistently recorded for the past 20 quarters.

Product revenues of $272.3m were down nine per cent from the year-ago $300.1m, while subscription services revenues grew 37 per cent from $96.2m to $131.4m. The company booked $724.8m in deferred revenue for the quarter, up 2.6 per cent Q/Q and 19.3 per cent Y/Y.

William Blair analyst Jason Ader told subscribers: “Despite the respectable quarter, revenue growth of 2 per cent represented a record low for the company, as the adverse impact of COVID-19 continues to weigh on the global IT spending environment in general and storage industry in particular.”

Wells Fargo senior analyst Aaron Rakers considers the deferred revenue increase “to be reflective of subscription expansion.” Pure’s total pipeline of opportunities for the third quarter is higher than in the second quarter but remains at an earlier stage, according to Rakers. He said this highlights the U.S. reopening and re-closing as negatively impacting spending decisions (relative to Europe re-opening and seeing better spending).

He told subscribers: “With clear pressures in on-premise infrastructure spending, we think Pure’s results will show relative outperformance vs. its storage peers.”  Rakers said Pure had seen consistent win rates and most notably competes against Dell EMC high-end PowerMax systems. 

In his prepared remarks, CEO and chairman Charlie Giancarlo said: “The current environment continues to challenge us all. … Q2 was a quarter of replanning for many organisations. Companies paused to consider the future and reassess their technology direction, aligning it to the new reality that their organisations face. Although the situation remains fluid, we believe the second half of the year will see businesses return to and even accelerate digital transformation.”

CFO Kevin Krysler said: “We are particularly pleased with the sustained strong growth and momentum of our subscription services that offer customers a cloud-like experience with more flexibility and compelling total cost of ownership.”

Chart graphically showing revenue growth slowdown in second 2021 quarter (green line)

Financial summary:

  • GAAP gross margin 68 per cent
  • Operating cash flow $50.7 million, up $1.9 million Y/Y
  • Free cash flow $25.7m, up $5.8m Y/Y
  • Total cash and investments $1.3bn

Pure added 350 new customers in the quarter, up from 300 in the previous quarter and 450 a year ago, and taking the total customer count to 8,150.

USA revenues were $282m, accounting for 69 per cent of total revenues and down 4.5 per cent Y/Y while international revenues were $124m, up 20 per cent Y/Y. Giancarlo said: “The growth and strength we saw from our international business in the quarter provides an early, but positive, indication of increased demands for our solutions when crisis levels created by COVID-19 decline.”

Pure is not providing formal guidance for the next quarter but informally expects that total revenue will be approximately flat sequentially. Giancarlo said: “I am confident in our opportunity, long-term strategy and ability to reaccelerate growth upon exiting the global crisis.”

NetApp layoffs hit Solidfire hard

By
Chris Mellor
-

NetApp is laying off up to 700 people, with Solidfire engineers taking a big hit.

More than 100 staff at a NetApp Solidfire facility in Colorado have lost their jobs, according to our sources. Overall, the impact has been greatest on field sales teams, with HCI (Hyper-converged infrastructure) and alliances also hit.

There are many anonymous comments on Thelayoff.com, some of which discuss a 5.5 per cent headcount layoff, and others a large layoff impact on Solidfire all-flash array business, which NetApp bought in 2015. For example, “Finally the trigger is pulled on a failed acquisition. The Solidfire acquisition was a total failure and has been known internally for years.”

According to one B&F source, NetApp sales reps tend to turn HCI sales leads into ONTAP sales, with some 86 per cent of HCI leads converted to NetApp filers.

.

Pure aims to kill hybrid storage with FlashArray//C

By
Chris Mellor
-

Pure Storage thinks the second generation of the FlashArray//C, launched today, is a hybrid flash-disk storage killer.

Updated 26 August 2020 with Matt Kixmoeller comment

The FlashArray//C is intended for tier 2 workloads such as test and dev, backup and recovery, VMs, email, data repositories and legal archives. With the new upgrade it costs less than enterprise hybrid arrays, Pure claims, and can host all enterprise performance-oriented and capacity-oriented use cases on a single all-flash platform.

Scott Baker, product marketing VP for FlashArray, said:  “We’re proud to … make the benefits of flash more accessible for a broader range of use cases at an economic advantage that is sure to make hybrid storage a thing of the past.”

Moving from enterprise disk drive arrays to its all-flash arrays can save multiple racks of disk drives and lower power and cooling costs, according to Alex McMullan, CTO, International at Pure Storage. He told us the array price is under $0.50/raw GB. That means a 1.3 PB gen 2 FlashArray//C would cost less than $650,166.50. On that basis, one with 31 drives and 768TB capacity would be priced under $384,000.

24.7TB DirectFlash QLC NAND drive

For comparison, a 768TB StorONE mirrored QLC flash (with Optane caching) array is priced at $594,489.00.  VAST Data also supplies a QLC flash array, again with Optane caching, and says it has a total cost of acquisition that is equivalent to hard drive-based archive systems. However, the company has published no actual prices. In common with Pure, VAST Data and StorOne argue that all enterprise storage can be consolidated onto a single all-flash system.

Pure’s Matt Kixmoeller, VP for Strategy said in an earnings call: “We now believe it’s substantially less expensive to deploy FlashArray//C than a hybrid array, about 30 per cent cheaper.” This could spark “the final transition to flash [to] happen.”

Spec

The gen 2 FlashArray//C uses higher-capacity NVMe QLC DirectFlash drives than its predecessor – 24.7TB capacity versus 18.3TB. Maximum raw capacity quadruples from 366TB to 1.8PB in 9RU, with effective capacity after data reduction also increasing fourfold from 1.3PB to 5.2PB.

The gen 2 arrays deliver 2ms to 4ms latency with on-drive SLC caching, but no Optane SSD caching, which helps keep the cost down.

The FlashArray//C will get 49TB drive modules later this year, doubling drive capacity and taking maximum effective capacity to 10.4 PB (2.6 PB raw) – eight times more than the gen 1 system. That will be the highest capacity mainstream flash drive. FlashArray// will get also more file storage features later this year, the company says.

The array’s Purity OS software manages quality of service and wear-levelling globally, with granularity down to the die level, to help ensure consistent performance and endurance. As a result, the 24TB drive is more efficient than an equivalent capacity Samsung drive with less over-provisioning, McMullan says.

Array reliability is helped by call-home telemetry down to the flash block level which enables Pure to run predictive analytics and help customers prepare for a drive failure by pre-emptively moving data.

Purity has policy-driven replication, snapshots, and migration between //X and //C arrays and clouds. Customers can use flash for application tiering, disaster recovery, test/dev, and retention. 

There is asynchronous replication from //X to //C but not synchronous replication.

FlashArray//C and FlashBlade

The Flash Array//C is positioned under the FlashArray//X, which has faster NAND chips and lower latency and is designed for tier 1 business-critical workloads.

In addition, Pure markets FlashBlade arrays for unstructured data, file and object storage, to optimise infrastructure for analytics purposes

We think there is some overlap between the enlarged FlashArray//C, with 5.9PB effective capacity and FlashBlade, with 3.3PB effective capacity in terms of the general analytics use case. The overlap will increase when FlashArray gets enhanced file storage features later this year. Perhaps the FlashBlade system could use the FlashArray//C drives and get a jump in capacity?

AWS cranks up EBS performance with io2 volumes

By
Chris Mellor
-
Cloud Love

Amazon Web Services has announced EBS io2, a new storage service that provides more durability and IOPs per GiB – at the same price – than the current io1.

Announcing the news today, AWS chief evangelist Jeff Barr writes: “The io2 volumes are designed to deliver 99.999 per cent durability, making them 2000x more reliable than a commodity disk drive … We are increasing the IOPS per GiB ratio … to 500 IOPS per GiB. You can get higher performance from your EBS volumes, and you can reduce or outright eliminate any over-provisioning that you might have done in the past to achieve the desired level of performance.”

Elastic Block Storage (EBS) volumes are attached to AWS compute instances. There are several kinds of EBS storage that differ in performance characteristics and price. SSD-based volumes are rated in terms of IOPS, and HDD volumes are measured in throughput (MiB/sec). With the new io2 volume the six types are;

  • General Purpose SSD (gp2), to 16,000 IOPS
  • Provisioned IOPS SSD (io1 and io2) for low-latency or high-throughput workloads, to a maxium of 64,000 IOPS.
  • Throughput Optimised HDD (st1) for streaming workloads, up to 500MiB/sec,
  • Cold HDD (sc1) for infrequently-accessed data, up to 250MiB/sec,
  • Magnetic (standard, a previous-generation type) supporting up to 90MiB/sec and 200 IOPS.

The io2 volume provides up to 500 IOPS per GiB – ten times for than io1’s 50 IOPS/GiB. It also has a 99.999 per cent (0.001 per cent annual failure rate) durability, better than the io1 volume’s maximum 99.9 per cent (0.2 per cent annual failure rate).

Barr says “io2 volumes [are] a perfect fit for your high-performance, business-critical databases and workloads. This includes SAP HANA, Microsoft SQL Server, and IBM DB2.” Apache Cassandra, MySQL, and PostreSQL can also benefit from io2 volumes.

Mohammad Shaikh, Director, Scientific Computing Services, Cloud Computing & DevOps, at Bristol Myers Squibb, provide a supportive quote: “With the 10x increase in IOPS per GB ratio, we can easily enable peak performance at a much lower cost than we ever could with traditional SAN hardware vendors.”

An AWS web page provides EBS pricing details.

Seagate readies 18TB IronWolf NAS drive for mid-September launch

By
Chris Mellor
-

Some online retailers are listing an unannounced 18TB Seagate IronWolf Pro NAS disk drive.

For example, Sweden’s  Data Byran is one emailer listing the drive, pricing it at 6,990 kronor ($798.36). A Reddit thread lists more etailers selling the dive and mentions mid-September availability. A Google search for “ST18000NE000” reveals many more.

IronWolf Pros are SME, SOHO and home user NAS drives with the current maximum capacity bring 16TB. The drives spin at 7,200rpm and have a 6Gbit/s SATA interface backed up by a 256MB cache and a 300TB/year workload limit.

Seagate started shipping enterprise data centre-class 18TB Exos drives to selected customers in June, and its technology is now trickling down to the IronWolf product line. 

Seagate has steadily increased the IronWolf Pro disk drives maximum capacity from 12TB through 14TB to 16TB. 

PowerStore vs Primera: Dell and HPE duke it out

By
Chris Mellor
-

Dell EMC-commissioned Principled Technology testers say PowerStore performs better than HPE’s Primera array while HPE, citing Dell EMC’s own numbers, says Primera performs much better than PowerStore.

Yes, we are also confused by this tit-for-tat exchange.

PowerStore bezel.

Principled Technology’s report says that, compared to the HPE Primera A670, the PowerStore 7000T has;

  • Up to 3.5x better data reduction
  • Up to 135 per cent more bandwidth
  • Up to 209 per cent more IOPS
  • Lower latency

The PowerStore 7000T is a 2-controller, all-NVMe flash drive array in a 2U chassis, offering unified file and block storage with always-on deduplication and RAID 5 data protection. The Primera A670 is a dual-controller external block storage array with a low-end 2U, 2-controller configuration. It is also all-flash, uses NVMe SSDs, and employs an ASIC for deduplication and other functions.

Primera models.

Principled Technologies used Vdbench to test block storage IO performance on the 7000T and the hyperconverged, VM-hosting 7000X, and the Primera A670. It tested IOPS, bandwidth and latency and you can inspect the full test details in a separate document.

We have ignored the 7000X results, as comparing a hyperconverged infrastructure appliance to an external storage array is not an “apples for apples”performance comparison.

Test results

The PowerStore 7000T “offered an overall 7.1:1 data reduction ratio compared to the 2.0:1 ratio of the HPE Primera A670.”

The PowerStore 7000T delivered 232,602 IOPS while the A670 provided 75,160 IOPS with data reduction turned on and 171,772 with data reduction turned off. The test used an 8KB random 100 per cent write workload. The PowerStore also performed better than the Primera on mixed read-write workloads.

Looking at bandwidth, the PowerStore provided 23,239 MB/sec on a random read workload compared to the Primera’s 10,763. Sequential read testing saw the Primera’s 9,877 MB/sec eclipsed by the PowerStore’s 23,238 MB/sec.

PowerStore’s latency was 0.5ms with an OLTP‐like mixed read-write workload comprised of 8KB and 128KB block sizes and a variety of read/write ratios at a fixed IOPS rate. The Primera’s latency was 2.01ms, four times longer.

Principled also looked at array snapshot restores, with a snapshot restore of 10 LUNs mounted as raw device mappings (RDM) onto a single VM. The PowerStore 7000T took a little over a minute, requiring just 16 steps. The Primera A670 array took over 20 minutes and 31 steps.;

Te testers concluded “PowerStore 7000 series arrays reduced data more efficiently and offered greater storage performance, as measured by IOPS, bandwidth, and latency” than the Primera A670; an apparently clear cut result.

Dell EMC is a SAP

HPE’s Dimitris Krekoukias, a global technologist and architect, delivered a robust response, citing Dell EMC’s own PowerStore performance numbers to show that Primera blows PowerStore away.

Krekoukias says he tried replicating the Principled Technology numbers but “I can’t come evenremotely close to replicating;any of their numbers.” Instead he looked at how many SAP HANA nodes the storage system can reliably support, using SAP’s certified system list.

Storage systems and SAP HANA node support.

A 2-controller A670 supports 36 SAP HANA nodes while a 4-controller one can run 72. The PowerStore 9000T, which has 112 controller cores compared to the 7000T’s 80, supports just 16 SAP HANA nodes. In other words the Primera A670 performs at more than two times better than the PowerStore 9000T when supporting SAP HANA nodes.

Krekoukias asks: “If the PowerStore is as fast as Dell claims in their FUD document, why isn’t it able to support many, many more SAP HANA nodes than 16?”

He points out that the Primera’s RAID 6 protection is also better than PowerStore’s RAID, and declares; “A Primera 670 is multiple times faster than a PowerStore 9000T [and 7000T], has stronger data protection, and much higher uptime.” 

B&F understands that the HPE EULA (Ed User License Agreement) doesn’t prohibit anyone from publishing performance results. The Dell and VMware EULA does.

Comment

Are the Principled Technology testers’ results wrong? We don’t know. This situation is why we need independent and verifiable external benchmarks which neither vendor references in their own performance statements. Don’t they want objective performance test results?

Still, it brings a smile to our face when HPE devalues Dell EMC’s sponsored test attacking Primera by referring to Dell EMC’s own SAP HANA certification tests – which show Primera is better in real-world scenarios. Dell EMC has been hoist by its own petard.

BackupAssist builds bridge between backup and DR

By
Chris Mellor
-

BackupAssist has released a product that bridges backup and disaster recovery.

The company’s eponymous software provides backup for Windows virtual and physical servers and applications such as MS Exchange and Office 365.

Linus Chang, BackupAssist’s founder and CEO, provided a quote: “We’ve known for some time that legacy backup systems are becoming less effective each day, as cyber-attacks increase in sophistication and target backup data in addition to primary data. To compound problems, the COVID-19 pandemic has meant that traditional backup procedures, such as swapping hard disks, are difficult or impossible during a lockdown.”

According to BackupAssis, local backups do not provide disaster recovery, and DR products are expensive and proprietary.

BackupAssist graphic.

BackupAssist new ER (Express Recovery) provides disk to disk to cloud protection, with local copies providing fast response and the cloud copy supplying the DR.

The company said the software is affordable because it uses low-cost S3 or Azure Blob storage, which is deduplicated and compressed, and has a recovery time measured in hour. Restore for the local copy take minutes. BackupAssist says the cloud recovery time is acceptable because disasters are rare, unlike local restores which can happen often. Backup data is encrypted and support includes AWS, Azure, Digital Ocean or any S3-compatible cloud, including self-hosted S3 clouds.

Restoring from the cloud can be via VM instant boot (booting a backup as a Hyper-V guest), bare metal server recovery or by downloading the backup to anywhere. Customers can recover entire applications or individual files.

BackupAssist DR recovery options

The company said its CryptoSafeGuard feature protects backups from ransomware in two ways. The software prevents unauthorised processes from creating, deleting or updating backup data. Also, it scans backup data for signs of a ransomware infection and blocks jobs from running. CryptoSafeGuard sends an SMS and mail to alert customers to detected risks. 

Chang said: “Our partners and end customers spoke and we heard – they wanted automatic offsite backups with history and retention, fast local recoveries, and the option to recover in the cloud when needed – all at a price an SMB could afford.”

BackupAssist ER can be bought online from BackupAssist or from its channel partners. It is priced at $299 annually per physical server, $199 per virtual guest machine annually or $699 per virtual host machine annually. There are discounts for bulk and multi-year purchases. The firm claims its pricing makes BackupAssist ER as much as 75 per cent cheaper than the competition.

The company

Just when you think you have a reasonable understanding of data protection suppliers along comes another one.

BackupAssist was founded in Melbourne, Australia in 2001. It is privately-owned, with no VC funding. The company has four products which are all versions of its core BackupAssist software: BackupAssist Classic; the new BackupAssist ER (Express Recovery and Expert Response); BackupAssist WFH (Work From Home) ;and BackupAssist 365.

The company says it provides affordable backup and disaster recovery software for small and medium business Windows Server customers. Its software protects more than 200,000 Windows servers.

The company supports physical and virtual servers and  has thousands of customers in 165 countries with offices in Australia, Malaysia, Europe, Africa and the USA. Larger customers include NASA, Holiday Inn, Siemens, Stanford University and Rolls Royce.

1...339340341...444Page 340 of 444

ABOUT US

Blocks & Files is a storage news, information and analysis site covering storage media, devices from drives through arrays to server-based storage, cloud storage, networking and protocols, data management, suppliers and standards. We publish news, views, opinions and analysis of storage industry events and developments, aiming to keep storage users, developers, consultants, marketeers, salespeople, support staff, analysts and everyone else involved in IT storage up to date with what’s going on.

FOLLOW US

© Situation Publishing, 2018-2024