We, Yoann Castillo and I, had a discussion about Quantum for our fifth Let's Talk for our Podcast Français du Stockage. We invite you to listen to it below but our talk is in French.
Google announced early this month the availability of Anthos for bare metal with a list of validated partners. It marks also the fundamental addition of environments without any hypervisor layer being the perfect layer to glue various operating platforms. It translates immediately these bare metal environments in the global Anthos initiative and offer them to benefit from this new giant leap in IT management approach. Anthos confirms its breakthrough and more generally the long term strategy of Google started many years with this vision. Just brilliant.
Robin Cloud Native Storage aka CNS is available on Google Cloud Marketplace. Confidential by its name but well respected for its technology and product, Robin tried an initiative to boost its adoption and market recognition. They created a FREE for ever solution full features just limited to 5 nodes and 5TB. Small configuration, this toy environment is perfect to test Robin CNS functionalities, stress the solution and potentially adopt it for larger configurations.
But things have changed and Violin evolved from a high-end positioning to a mid-range one before that move. StorCentric continues this and adds Violin to the Nexsan category. The shift was big as today Violin relies on standard x86 servers with storage software, so very similar to other approaches at a macro level. That delivers a real rich portfolio against usual competition. The challenge for StorCentric would be to consolidate software and use common elements across the product line. There is some directions to add some file access methods to Violin to become an unified product like Unity but this time as a full flash one.
On the software side, beyond data management, the company insists on security and especially ransomware to prevent business impact leveraging Retrospect intelligence.
Being always ready for a bargain, Mihir Shah and his team, plans more acquisitions to continue to build a serious comprehensive storage portfolio. Historically very block storage oriented except with Drobo and Nexsan Unity, the team has a clear software strategy both in terms of Software-Defined Storage and also in more general data management aspect of it. To support this, the firm has released a few weeks ago Data Mobility Suite aka DMS for all kinds of data movements especially in heterogeneous storage environments. But we see also some directions with the wish to acquired some technology in data analytics for data and storage management and an object storage engine.
Back to DMS, the idea is to provide a generic copy and replication engine among StorCentric products but also between with other products. Four use cases are covered here: data migration with the need to reduce cost or refresh systems, data replication to enhance disaster recovery and potentially business continuity, data synchronization across various storage entities and finally multi-site collaboration. A story to illustrate how to leverage an universal copy engine with different use cases. We'll monitor the adoption of DMS as supported cloud providers are interesting even if some of are completely obsolete.
The recent session during The IT Press Tour confirmed some developments around NVMe over Fabric and the preferred transport layer is TCP and the wish to add an S3 access method as well.
Last week, the team has made an new TrueCommand iteration and unveiled version 2.0 during a new edition of the press tour, #37 this time, with some cloud capabilities. The product is named TrueCommand Cloud and controls any TrueNAS flavor - Core, Enterprise or Scale - running on-premises as a VM or a Docker container or in the cloud in AWS. Features got extended supporting more TrueNAS features with a self-service portal and security built-in.
As said multiple times, MinIO Tsunami is here.
Recognized as the obvious leader in HPC storage, DDN shipped 10+EB since its inception which means a lot and delivers approximately $600M annual revenue confirming as well its unicorn status.
The company has been also elected as a leader in the annual file storage Map 2020 by Coldago Research. You can access the report here.
Let's dig first in the "At Scale" business. The first key strategic decision taken a few years ago now is the abandon of IBM Spectrum Scale or GPFS in favor of Lustre, the open source distributed file storage software. DDN has acquired Whamcloud in 2018, the Intel business unit dedicated to HPC file Storage absorbed by Intel in 2012. In 2018, DDN sold probably 50/50 between Lustre-based storage named ExaScaler and GPFS-based named GRIDScaler but in 2020, 4 ExaScalers are sold vs. 1 other product representing the rest of the HPC product line. This is just impressive. During the recent months the product team has delivered a set of key things paramount for DDN market success: A3I, dataFlow and of course ExaScaler 5th generation.
AI is closer to HPC than we think and it triggers some needs in I/O, capacity and bandwidth requirements. It explains why ExaScaler is a good fit but also why some additions were added to the product:
DDN occupies a unique position in the market explaining why its collaboration with Nvidia shows strong integration.
DDN ExaScaler supports real multi-cloud experiences being available on-remise but also on AWS, Azure and GCP. Dashboards, GUI, realtime monitoring and analytics made its product really compelling this year illustrated by a good traction.
At the same time, Exa5 is positioned as the unstructured product if choice with a very comprehensive in the file and object side from an access method perspective.
This convergence I affirmed for several years now is what DDN finally promoted as well. Enterprises needs HPC and HPC considers enterprise techniques as well, being reciprocally beneficial.
DDN also develops a new multi-cloud data services infrastructure that should be available in June when The IT Press tour will visit again DDN in California. Perfect for us.
StorPool is a real SDS if you consider my definition: "SDS transforms a rack of (classic and standard) servers (with internal disks) into a large storage farm. You can find on the market block, file or object storage as it is essentially how the storage entity is exposed outside". In other words, take 3 servers, install Linux and StorPool software and you get an high performance and resilient storage array.
The team develops probably one of the most comprehensive distributed block software layer, others are ScaleIO, acquired by EMC, Datera, Kaminario/Silk and more recently Lightbits Labs or Excelero to list a few.
With a minimum if 3 nodes, the team implements a very rich block storage offering:
In a nutshell, the product delivers outstanding performance levels, look at these numbers:
One element that continues to surprise me is that StorPool doesn't support erasure coding but just replication. Replication is a good mechanism, no doubt about it, but erasure coding at large scale and large files is also interesting. It has demonstrated some huge impact and some others players offer this. So it invites me to think about a potential difficulty in their design to add EC or they don't receive such requests. But why? as value is better durability, better storage size efficiency and lower costs. Let's do simple math to compare. If you consider 500TB of usable storage, 3 copies mean 1500TB of raw storage with an overhead of 200% (x3) and an efficiency ratio of 33%. If you compare with EC with 6 data fragments and 2 parities fragments, noted EC 6+2, you obtain 33% overhead and 75% of efficiency. Of course I can even add one more parity to reach 6+3. So it means we can do almost similar durability with just 667TB or 42 HDDs. Let's put price in the equation and consider a common 16TB SATA internal drive on Amazon at $400. The difference in capacity is 833TB (500 x 3 - 500 x 1.33) which represents 52 HDDs and $20,800. And here we don't count the energy, the chassis size or number of nodes... as we reduce by 52 globally the 94 initial drives number (1500 = 94 x 16). The total 1500TB costs here $37,600 dropped to $16,800. With that StorPool would become unbeatable...
The team focuses on performance and achieves impressive numbers available publicly with different benchmark tools and environments. It explains why the adoption is started for a few years confirmed by the profitability of the company.
As a business, StorPool sells via a mix of direct and resellers and the French ISP Iguane Solutions is a customer for instance. This cloud and internet providers represent a key strategic direction for the company for several years and the footprint there is significant.
We'll continue to follow StorPool, again one of the key European storage innovator in block SDS.
2020 was a very good year for the company showing strong growth despite the Covid pandemic. Henri Richard joined the board of directors, Data Dynamics acquired Infintus for content indexing and search strengthening its data governance aspect and Lenovo partnership skyrockets.
StorageX 8.3 was released recently as a minor release essentially with improved performance. The product continues to be the preferred choice for data management for scalable NAS for everything related to data replication, distribution, collaboration, protection or optimization of cost and storage. Recent developments were made around object and cloud storage therefore S3 and Azure Blob to integrate seamlessly in the data movement workflow.
StorageX had demonstrated for more than a decade its unique capabilities illustrated by NetApp and Lenovo oem deals.
Data Dynamics insists also on the platform notion of StorageX as a major differentiator against other market tools or products. A tool is just for very limited task during a limited portion of time and then you forget it. A product is better in terms of usages, it is more larger and can reside longer but it's still limited to very limited usages. Therefore to deliver consistent regular efficient data services and more globally IT services, administrators have to multiple tools and products and by effect creates obviously some complexity that increases costs and degrades performance and efficiency finally. A platform is a radical new and different approach, it is resident, you build services around it and therefore it structures how administrators manage services as it delivers multiple key functions from the same solution. Its role is central and paramount and used everyday representing a strategic IT service asset to sustain business.
The StorageX companion is Insight AnalytiX built on the Infintus content indexing and analysis engine to extend data management towards data governance with regulations alignments and compliance and therefore contribute to limit risk exposure.
2021 should be a new year of market success and penetration for Data Dynamics, we expect some revenue progress but also new partnerships especially around hyperscalers.
