Tuesday, June 25, 2019

New dense appliances for Qumulo

Qumulo, one of the NAS market leader, announced today several new appliance models and software enhancements.

Three new models are introduced confirming a real trend towards dense nodes with NVMe drives. The Performance model P-184T receives double the capacity of the existing one with 184TB in 2U now with 24 drives of 7.68TB delivering 225GB/s with 4 x 100GbE ports, the hybrid C-168T has now 12 x 14TB HDDs coupled with 4 x 960GB SSD in 1U with 2 x 25GbE ports dedicated for mixed workloads and the active archive K-168T mimics the hybrid new storage update using 2 x 10GbE networking ports.. New software enhancements also boost performance by 40%, it is available to currents models with a simple OS upgrade.


Qumulo also updates the analytics engine for capacity consumed by storage, snapshots and metadata. Following the XPP introduction a few weeks ago, a security activity audit is added to understand information about file operations and activities.

Qumulo confirms its leadership with a very comprehensive solutions for cloud and on-premises workloads and demanding file environments in performance and capacity. The advantage to have designed an internal distributed file system creates a real gap with other NAS vendors. Next few months we should see a real acceleration and potentially an IPO.
Share:

Monday, June 17, 2019

A few good deals to secure

Cloudera last trading value was around $5 Friday, Nutanix touched $25 and Pure Storage landed on $15... to list a few bargains. Even MapR closed to a shutdown belongs to this list failing to raise a new VC round.

Cloudera would be great at IBM to increase Hadoop and services portfolio around Big Data and even HPE always looking for good deals after SimpliVity, Blue Data and Nimble Storage.

Who else? Hum I see some commercial object storage vendors who tried to survive and swim in a red ocean with too many players and reduced opportunity especially for pure players. As S3 eats everything, Minio became ubiquitous and the Tsunami I announced already a few years ago for S3 and open source object storage is confirmed. Les start a countdown for others, I don't see any future for several commercial others...
Share:

Wednesday, June 12, 2019

iXsystems unveils TrueCommand

iXsystems, a leader in systems and storage based on open source software, just introduced TrueCommand 1.0, a centralized management console for TrueNAS and FreeNAS product line deployed across the globe.


It offers performance monitoring and reporting, fault diagnosis, ZFS aware with specific integration, predictive analytics for capacity and health. By ZFA aware we mean the TrueCommand understands ZFS pools, caches, datasets and snapshots.
The tool is free of charge up to 50 drives not systems and enterprise licenses are available for larger configurations. You need a x86 64-bit CPU, at least 4GB of RAM and 80GB of disk space. TrueCommand is available to download via this link.

Share:

Monday, June 10, 2019

Zero compromise guarantee for VAST Data

VAST Data, emerging leader in All Flash NAS, continues to shake the industry this time with what they called Zero Compromise Guarantee to promote their high availability, durability and longevity of their infrastructure. I wrote a post when they unveiled their product a few months ago.

Thanks to real new approach in data reduction, placement and protection, the company is firmly confident that QLC/Flash will last 10 years. They even promote a support and maintenance contract at 10% meaning that reaching 10 years you just pay double your solution at that. In the industry we can see 20 even 23% maintenance contract meaning that in less than 5 years people pay double the initial public price.

This is clearly a second key argument beyond performance of VAST Data's Universal Storage platform. IT also means that the company is ready to innovate not only on the technology side but also on the business side with a new relationship with their end-users and prospects.

In the announcement, the company insisted on a few key elements inviting implicitly prospects and even users to ask competitors their behaviors in these areas:
  • No data loss as data are always written to durable storage,
  • No downtime as controllers are redundant and can serve and access same data,
  • Advanced data reduction that contributes to the longevity of flash drives,
  • All features in the product i.e no pricey options,
  • and of course erasure coding at large scale to protect data over long run and guarantee the persistence of data.
The product introduces definitely some new level of availability and durability needed at scale for numbers of new use cases in current digital economy.
Share:

Friday, June 07, 2019

WekaIO breaks a new record

Benchmark results are key when you address technical computing and especially HPC. It exists essentially 2 flavors of benchmark: the ones to beat records and most of the time at a very high price and only a few entities can afford them and ones on users' hardware to beat performance thresholds on more common deployments.

We met WekaIO at their new HQ 3 days ago with The IT Press Tour and we got a very good session as usual following the visit we did at their R&D center in Tel-Aviv in November 2016, almost 3 years ago, the company has made lots of progress.

WekaIO has chosen to publish regularly various performance benchmarks on different hardware platform like SPECsfs and more recently did the STAC-M3 Antuco and Kanaga test on the FrostByte and Relion systems made by Penguin Computing.

This test uses the Kx time-series distributed database kdb+ 3.6 deployed across 7 Relion servers with data stored on FrostByte storage with 9 NVMe SSD per 1U server, Mellanox router and adapters with WekaIO MatrixFS 3.2.2 as displayed below. The result is impressive with 37.5GB/s and 2.5M 4K IOPS.

MatrixFS has a new design addressing limitations of "classic" distributed file system such Lustre and Spectrum Scale. It is what we called a Symmetric Parallel Distributed File System: Symmetric means that metadata service is delivered by any server in the cluster, Parallel implies that clients distributes/chunks/splits/stripes file data among multiple data servers thanks to the agent loaded on each client system and Distributed refers to the file system layout existing across data servers.

As financial demanding applications require high performance, financial institutions consider seriously such results that serve as a very good vehicle to boost WekaIO customer adoption.


But there is a mystery and just a miss: WekaIO is listed on the Penguin partners page but the FrostByte product page continues to list BeeGFS, Red Hat Gluster and Ceph and Lustre and no WekaIO as I wrote last year. Bizarre.

Share:

Thursday, June 06, 2019

Tuesday, June 04, 2019

The unstructured data catalog from Data Dynamics

A good surprise of the IT Press Tour this week comes from Data Dynamics who markets StorageX, the famous Network File Management and Virtualization product born at NuView then acquired by Brocade in 2006 for $60M.

Data Dynamics continues to penetrate enterprise landscape with 26 Fortune 100 companies, 6 of 12 world's largest banks and operating 350PB+ over 17 years of technology market presence. It means that the technology is reliable, proven, well adopted and span multiple technology waves and refreshes. We realize that StorageX lasts longer than the storage it manages and for a simple reason data lifecycle is longer that the hardware device where it resides.

We had the privilege to meet Piyush Mehta, CEO of Data Dynamics, now owner of StorageX since 2012, and got an interesting update on the product and future directions.

About the product, StorageX is one of the most comprehensive file data management and analytics platform based on a very advanced metadata engine. However it suffers from a lack of visibility on the market.


It operates as a global file directory service for on-premises and cloud file servers and NAS like DNS plays for Internet. The slide above illustrates the features set of the product around the analysis of the unstructured data environment.

StorageX leverages a SQL database for several years, StorageX integrates now Elastic search and NoSQL engine when analytics and archives modules are enabled, for storing meta data collected and the tagging that drives categorization and intelligence into the data. This information can be accessed via StorageX UI, API or directly via Elastic.

In term of deployments, StorageX server receives data from UDE - Universal Data Engine - collectors running on-premises. As a pure software, StorageX doesn't require agents, gateway or stubs.


With the 8.1 release, several new features were introduced:
  • Analyze: optimized file scan and analytics powered by Elasticsearch plus the capability to perform query search to create custom datasets for effective data analytics and reporting,
  • Move: support added for Microsoft Azure Blog meaning that StorageX adds object storage target,
  • Manage: Integration added for 3rd-party monitoring tools plus enabled data grooming for Cassandra,
  • Modernize: New API functions for business workflow integrations.
StorageX is also a key solution chosen and selected for the delivered proven ROI. The image below shows an example that illustrates perfectly generated savings.


The pricing model for all StorageX modules is based or capacity per year per account in addition to a StorageX engine fee.

We expect StorageX 9.0 soon with some interesting new features and product architecture. First in term of architecture, Piyush Mehta told us that the product will be redesigned with microservices for the core and edge i.e the data collectors. AI will be also included and security extended. The last addition will see content analysis to complement metadata knowledge, enhance policy engine and align with some compliance requirements. Very good session, hope to meet Data Dynamics for a future IT Press Tour to see progress.

Share:

DataCore ready for a new era

We met a few times DataCore several years ago at heir HQ in Fort Lauderdale and the company seemed to manage pretty well its installed base but had some difficulties to go beyond its storage virtualization comfort zone.

The meeting early today during The IT Press Tour was the opportunity to meet the new team, at least part of it, especially Dave Zabrowski, CEO, Gerardo Dada, CMO and Steven Hunt, Product Management.

But a sad event has changed completely DataCore. The death of Ziya Aral, Founder and CTO, and technical hero of the company, early 2017 was a shock and served as a real trigger for the company to imagine a new future wishing to leverage its strong recognized expertise for two decades. Ziya co-founded DataCore in 1998 with his close friend George Teixeira, now chairman of the company.

After several months of shaking heads, the new DataCore landed on earth again. As you may have seen for severals months, the company name is the same but the logo, mission, strategy and management have changed under the leadership of Dave Zabrowski, arrived in April 2018.

The company has made a serious exercise to study end-users and partners needs especially their customers to understand their expectations and above all consider them. First, a clear cloud strategy is missing as DataCore is an on-premise player so the public flavor was out of their scope. Does it mean they will announce something, we'll see. Considering this domain, the pricing strategy will require an change to match the now common practice which is a subscription model.

The second direction could be to leverage the SANsymphony SDS software and start an HCI offering. This needs a hypervisor choice and other software adherence.

We then expect of course a new iteration for SANsymphony and what about what a few others developed overtime like Nimble Storage InfoSight, Kaminario Clarity, NetApp OnCommand Insight or Pure Storage Pure1. I mean a SaaS Analytics an Insights platform to understand end-users' storage environment getting more and more complex and large. We'll see but we heard that DataCore have some plans on these few product directions.
Share: