Technology:Unpacked

This week, Chris and Martin debate whether we have seen the end of new “pure play” storage companies that go the distance to full independence. Where there used to be many businesses like EMC, Pure Storage and NetApp, the number of public storage-only companies is dwindling. Is this market too challenging to get into or is there simply no money to be made?

If we look back 30 years, storage hardware was all the rage. Storage solutions were complex with custom hardware. These days, anyone can build a software-based solution and open source for everyone to use. We’ve also seen many companies raise hundreds of millions of dollars and be in no rush to IPO. So what’s going on? Listen to find out.

Vendors mentioned in this podcast: Pure Storage, NetApp, IBM, NGD Systems, Seagate, Western Digital, Tegile, Infinidat, VAST Data, Scality, Cloudian, Rubrik, Cohesity, Veeam, Nutanix, StorONE, Pavilion, DDN, Hammerspace, Storpool, StorageOS, Portworx, HPE, Hitachi, Oracle, Dell EMC, Softiron, MongoDB, Postgres, Databricks, Nebulon.

Elapsed Time: 00:43:11

Timeline

* 00:00:00 – Intros * 00:03:20 – Storage hardware used to be cool * 00:06:00 – Unicorns and hectacorns! * 00:06:30 – Are companies looking to IPO or be acquired? * 00:07:50 – Margins on storage are thin * 00:11:15 – The Next Wave – Seagate/Western Digital * 00:14:10 – Infinidat – building traditional HDD systems * 00:17:30 – We like Howard! * 00:20:00 – Companies need new product lines * 00:21:00 – Where are object storage companies headed? * 00:23:45 – What about data protection companies?  * 00:26:00 – Secondary data usage means working with the business * 00:32:00 – Some companies will need to merge * 00:35:00 – Buying startup solutions could be risky if the companies are acquired * 00:36:30 – Open source storage could be an attractive fallback * 00:40:00 – Wrap Up

Related Podcasts & Blogs

* #168 – Storage Unicorns* #130 – Making Money in the Storage Business* #138 – Storage Predictions for 2020 (Part I)* #139 – Storage Predictions for 2020 (Part II)

Copyright (c) 2016-2020 Storage Unpacked. No reproduction or re-use without permission. Podcast episode #hpno.

This week Chris and Martin talk to Nikita Ivanov CTO and founder of GridGain Systems. The topic is in-memory computing and specifically Apache Ignite, an open-source key-value store that also supports SQL99 and POSIX-compliant file interfaces.

The idea of running applications purely from memory isn’t a new one. DRAM is the fastest “storage” component but isn’t designed as a long-term storage medium. Consequently, in-memory solutions such as Apache Ignite require features to ensure data resiliency and consistency. Ignite and similar solutions have a heavy focus on data distribution and protection in order to meet resiliency needs.

We also have to remember that memory and storage use different semantics. Memory is byte-orientated, through LOAD/STORE type instructions, whereas storage operates at a block level through read/write introductions. This difference provides both opportunities and challenges. As Nikita indicates, the new wave of storage-class memory products (persistent memory) such as Optane may seem to offer benefits, but might not offer significant benefit through the addition of persistence.

You can learn more about GridGain at https://www.gridgain.com/ and Apache Ignite at https://ignite.apache.org/

Elapsed Time: 00:45:35

Timeline

* 00:00:00 – Intros * 00:01:10 – What is Apache Ignite? * 00:02:30 – Effective in-memory computing introduces multiple machines & distributed systems * 00:06:20 – Memory and storage have different access semantics * 00:09:00 – In-memory computing has driven the most advanced distributed systems * 00:10:24 – What data models does Apache Ignite support? * 00:12:00 – Ignite offers SQL99, Key Value and POSIX file system semantics * 00:13:19 – Ignite suits between 8 and 64 nodes * 00:16:00 – Ignite is aimed at high-end in-memory requirements * 00:18:21 – Is in-memory computing a replacement for faster hardware? * 00:22:30 – GPUs offer the ability to manage small-scale analytics * 00:23:50 – How can we differentiate between in-memory solutions? * 00:25:00 – Complexity is a challenge for in-memory computing * 00:27:30 – Do we need to modify in-memory computing to be more consumable? * 00:32:10 – How do we differentiate between the multiple in-memory solutions? * 00:34:00 – How will new media influence in-memory development? * 00:39:00 – The next challenge for non-volatile media is integration * 00:40:30 – Wrap Up

Related Podcasts & Blogs

* #147 – Introduction to Key Value Stores and Redis* #36 – The Persistence of Memory with Rob Peglar* #159 – Introduction to MRAM with Joe O’Hare from Everspin

Copyright (c) 2016-2020 Storage Unpacked. No reproduction or re-use without permission. Podcast episode #25c4.

This week Chris and Martin review the idea of storage unicorns, companies that have a valuation of one billion dollars or more. What exactly is the basis or justification for a billion dollar price tag? Is this something invented by the VC industry or is there a real degree of science behind the assumptions? The list in question comes from a Blocks & Files article written by Chris Mellor, which in turn references the list produced by an analyst firm.

While valuation based on some multiple of money invested does give some indication of value, there are other more tangible metrics available. Unfortunately the industry isn’t keen on sharing revenue and customer numbers other than when legally required to so.

Who is missing off the list? We see no mention of Weka, Scality, Cloudian, Wasabi or Backblaze. It seems unlikely these companies don’t have equivalent valuations, so perhaps the building of lists is more arbitrary.

You can find the article written by Chris Mellor here – https://blocksandfiles.com/2020/07/15/coldago-storage-unicorns/ – so make your own mind up!

Elapsed Time: 00:34:41

Timeline

* 00:00:00 – Intros * 00:01:00 – Chris Mellor’s article on Coldago Storage Unicorns * 00:02:40 – Unicorns have valuations greater than $1 billion * 00:04:00 – Datrium acquired by VMware, perhaps $600 million? * 00:06:00 – $1 billion is a lot of money! * 00:07:20 – Valuations can have no relation to the revenue of a company * 00:09:40 – How many storage companies have successfully IPOd in the last 5 years? * 00:11:30 – Where are Weka, Scality, Cloudian? * 00:12:40 – DDN is now a portfolio company (Tintri, Intelliflash, Nexenta acquisitions) * 00:16:30 – Are we over-cynical, or is it just opinion? * 00:19:00 – Companies need to have relevance to the public cloud * 00:20:05 – Party time!  Rubrik & Veeam compete – what will VMworld do this year? * 00:22:45 – Veritas – the oldest unicorn in the world? * 00:25:00 – Do the unicorns need to expand their product portfolios? * 00:27:15 – Will we continue to see storage unicorns? * 00:30:10 – Are we too Western focused?  * 00:31:00 – Who else is missing – Wasabi?  Backblaze? * 00:33:00 – Look out for the Storage Unpacked Storage Unicorn List! * 00:33:40 – Wrap Up

Related Podcasts & Blogs

* #130 – Making Money in the Storage Business* #110 – Storage Vendor Consolidations & Acquisitions* #64 – Success & Failure in Storage Startup Land

Copyright (c) 2016-2020 Storage Unpacked. No reproduction or re-use without permission. Podcast episode #p2fg.

This week Chris and Martin catch up with Phil Straw, CEO at Softiron for a discussion on packaging Open Source storage solutions for adoption by the enterprise. Softiron sells a software appliance based on Ceph which has been tuned to deliver high performance with high efficiency. So, exactly how do vendors go about making Open Source more consumable without breaking the values of Open Source software development?

This discussion is interesting as it highlights both the benefits of marrying hardware and software, while working within the constraints of a community software solution. Rather than simply test, Softiron look at the code and work out the right way to make Ceph work efficiently. Of course, every enterprise could do the same thing, but typically this task was covered by the vendor.

You can find more on Softiron at www.softiron.com.

Elapsed Time: 00:36:09

Timeline

* 00:00:00 – Intros * 00:01:30 – How can we package Open Source storage software? * 00:02:30 – Is there a gap in open source and enterprise expectations? * 00:06:00 – Where do Open Source storage ideas come from? * 00:07:30 – Softiron brings an “enterprise” experience to Ceph * 00:08:45 – Open source adoption often means just testing * 00:11:40 – Is the enterprise capable of managing open source? * 00:13:00 – How do you adopt a software product with little code control* 00:16:00 – HyperDrive is built to be a storage solution * 00:18:15 – Ceph is not forked* 00:20:00 – How can Softiron ensure Ceph heads the right direction * 00:24:30 – Storage is becoming more important again * 00:28:00 – Lock-in occurs in many ways, not purely technological * 00:29:00 – How can a vendor like Softiron protect their IP? * 00:32:45 – Always ask how you will exit using a product * 00:34:30 – What is HyperDrive? * 00:35:10 – Wrap Up

Related Podcasts & Blogs

* #41 – Does Open Source Have a Place in Storage?* #46 – Another View on Open Source Storage with Neil Levine* #130 – Making Money in the Storage Business

Copyright (c) 2016-2020 Storage Unpacked. No reproduction or re-use without permission. Podcast episode #504n.

This week, Chris and Martin are joined by Chris Mellor from Blocks & Files and Eran Brown from Infinidat to discuss flexible pricing strategies in light of current business disruptions. Infinidat has introduced an elastic pricing model for both capex and opex purchases. How is this implemented and what platform features enable Infinidat to deliver this capability?

As businesses are disrupted through the coronavirus pandemic, uncertain times means uncertain demands and budgets. Companies may defer spending or want flexible purchasing models for technology deployments. Public cloud is one choice, but this comes with challenges around data repatriation.

Infinidat uses low-cost media and proprietary data management techniques to optimise I/O across DRAM, disk and flash. With the majority of data on cheap storage, Infinidat is able to over-provision and offer customers much greater flexibility in purchasing options, including capital and operational expense.

Eran takes the team through the challenges, explains some of the issues customers have experienced and then details how Infinidat can offer flexible terms to customers over the lifetime of their relationship.

For more information on Elastic Pricing, check out https://www.infinidat.com/en

Elapsed Time: 00:44:28

Timeline

* 00:00:00 – Intros * 00:01:10 – Uncertain times, uncertain demand and supply * 00:02:30 – Budgets are under constraint * 00:04:30 – The storage industry is in a strong position * 00:06:00 – Is data centre access a challenge? * 00:09:00 – Are uncertainties driving businesses to public cloud? * 00:11:00 – Over-provisioning is one solution to avoid friction of purchase * 00:13:00 – Moving temporarily to the cloud has business implications * 00:17:00 – Private and public cloud purchasing are inverted * 00:19:00 – Flexible pricing allows customers to get best value from budgets * 00:20:00 – Can all-flash vendors afford to put over-provisioned systems onsite? * 00:24:00 – What is Infinidat Elastic Pricing? * 00:29:00 – Deploying to multiple customer sites is a complex process * 00:32:00 – Tiering to cloud is an impractical solution * 00:33:50 – How can AIOps and analytics help predict capacity demand? * 00:37:30 – Minority Report for storage purchases * 00:38:47 – Welcome to the Storage Bank * 00:40:00 – Purchasing and consumption models will be more important than speeds & feeds * 00:43:10 – Wrap Up

Transcript

Related Podcasts & Blogs

* #141 – Building Storage Systems of the Future* #132 – Accelerating Ransomware Recovery with Eran Brown

Copyright (c) 2016-2020 Storage Unpacked. No reproduction or re-use without permission. Podcast episode #m3km.

As we move to a hybrid and multi-cloud world, IT organisations need standardisation in operational processes. This is particularly true for data protection, where consistent policies and compliance are essential. This week, Chris chats to Subbiah Sundaram from HYCU about data protection in public clouds and the ability to bring a consistent reporting model with HYCU Protege.

Why is consistency important? Public cloud providers have relatively basic data protection options, that might not align with existing enterprise requirements. As workloads move around, backup needs to be consistently applied, whether that data is on premises or in a platform like Azure.

There are three choices; run multiple platforms and hope they can be manually aligned. Alternatively, SaaS offers great centralisation but doesn’t provide local recovery. A third option is to use backup software designed for each platform and a central management/reporting platform.

HYCU now offers data protection for on-premises, Azure and GCP workloads. Management reporting is centralised using Protege. Customers can also use Protege for application mobility, moving backups between platforms with consistent restore capabilities across clouds.

To learn more about Protege and HYCU for Azure & GCP, check out https://www.hycu.com/ and for Test Drive – find out more in this blog post. You can learn more about HYCU for Azure in this Architecting IT blog post.

Elapsed Time: 00:31:08

Timeline

* 00:00:00 – Intros * 00:01:00 – What is hybrid and multi-cloud data protection? * 00:04:30 – Moving to cloud needs a rethink of data protection * 00:05:05 – Cloud backup solutions aren’t very mature * 00:06:06 – Portability and multiple copies are still important  * 00:09:00 – Three scenarios – SaaS backup for everything * 00:09:30 – Scenario 2 – do nothing * 00:10:30 – Scenario 3 – centralised management distributed backup * 00:12:10 – Local copies enable fast recovery * 00:13:30 – Application mobility hasn’t really happened * 00:14:20 – Networking is still a major issue in application mobility * 00:19:00 – HYCU for Azure, how does it work? * 00:22:00 – What is Protege? * 00:27:00 – What about Futures – Containers?  AWS? * 00:27:30 – Nutanix Mine and HYCU Test Drive * 00:30:00 – Wrap Up

Related Podcasts & Blogs

* #73 – HYCU – Data Protection for Hyper-converged Infrastructure* HYCU Announces GA of HYCU for Azure* Data Protection Choices for Heterogeneous Environments

Copyright (c) 2016-2020 Storage Unpacked. No reproduction or re-use without permission. Podcast episode #vued.

Storage startup StorONE has introduced the S1: All-Flash Array.next (also known as S1:AFAn) which accelerates performance using Intel Optane technology. In this podcast, Chris and Martin discuss the benefits of Optane within the S1 platform in a conversation with George Crump, CMO at StorONE and formerly the founder of Storage Switzerland.

S1 is a software-defined storage solution that optimises the use of hardware resources to deliver an efficient and low-cost multi-protocol shared storage platform. The architecture of S1 enables Intel Optane to be used as a tier of storage that acts as the primary target for write I/O. Optane is used in S1 in conjunction with Intel QLC 3D NAND SSDs to provide both a cost-effective storage implementation and one where low latency and high throughput is achieved with only a small number of Intel Optane drives.

George explains the features of Optane, which include high endurance and low latency. A single S1:AFAn with just 3 Optane drives can deliver 1 million IOPS and up to 300,000 sustained write IOPS.

StorONE has taken the stance of providing pricing transparency. Customers can pre-configure systems using TRUprice, an online configuration tool that shows configuration pricing for multiple vendor hardware configurations or simply software only.

You can learn more about S1:AFAn at the StorONE website (https://www.storone.com/product/s1-afan/) and find more details on TRUprice here (https://www.storone.com/truprice/).

Elapsed Time: 00:40:34

Timeline

* 00:00:00 – Intros* 00:01:27 – George moves to the dark side!* 00:02:18 – What is the StorONE storage platform?* 00:03:45 – S1:AFAn Intel Optane Array* 00:05:19 – How is SCM/PM Implemented in S1?* 00:08:12 – S1 used Optane as a tier* 00:11:00 – S1 is a software-defined storage solution* 00:13:11 – Optane and S1 has significant latency improvements * 00:14:30 – Faster storage means more efficient CPU utilisation* 00:16:00 – Data hits the Optane tier first and is then archived down * 00:19:30 – Customers can implement flexible hardware configurations * 00:21:30 – Read performance from QLC is almost as good as Optane * 00:23:40 – Are we seeing the emergence of a new hybrid architecture? * 00:28:00 – Hero Numbers – how does Intel improve performance? * 00:31:00 – S1 also offers object storage through the S3 protocol* 00:35:00 – How is S1 priced?  TRUprice* 00:36:30 – Will StorONE help expose better cost transparency?* 00:38:36 – Wrap Up

Related Podcasts & Blogs

* #141 – Building Storage Systems of the Future* Managing Massive Media* The Expanding Storage Hierarchy* What are SCM and PM?

George’s Bio

George has over 25 years of experience in the storage industry, holding executive sales and engineer positions. Before joining StorONE, he was the founder and lead analyst at Storage Switzerland.

Copyright (c) 2016-2020 Storage Unpacked. No reproduction or re-use without permission.

Technology refreshes are a constant part of the IT landscape. Businesses need to frequently update and modernise their infrastructure. Typically, the inertia to move between vendors is strong, but inflection points do occur, when the cost and risk of change is outweighed by the benefits. How do you identify inflection points in your environment and then take advantage of the opportunity?

In this podcast, Chris chats with three guests (Matt Watts from NetApp, Lee Nolan from MTI and Kam Panesar from Atradius) to get views from a vendor, solutions provider and end user perspective.

The biggest hurdle to making a change is overcoming risk. Picking a new vendor has implications on costs, training and Day 2 operations. However, not making a change can be equally as impactful if the incumbent vendor changes strategy or has a confused portfolio of products.

In this podcast discussion, we cover RFPs, the effect of Public Cloud and how new technology is impacting the structure of IT organisations. In the end, technology is there to run the business and picking the right technologies is a constant process.

Elapsed Time: 00:41:39

Timeline

* 00:00:00 – Intros * 00:02:45 – What is generating the pace of change and inflection opportunities * 00:04:45 – What are the reasons for change? * 00:06:30 – Automation & DevOps is a current strategy * 00:08:15 – Not changing is a risk as much as change * 00:09:30 – How willing will your vendor be to help when problems hit? * 00:10:45 – Never take a new product until version 3! * 00:14:00 – Vendors have inflection point risks too * 00:15:30 – Is there still value in the RFP process? * 00:22:00 – Are we skilled enough to translate business into technical requirements? * 00:25:45 – Is Public Cloud sexy? * 00:28:00 – Mature, stable workloads probably won’t benefit from migration to cloud * 00:29:30 – Data mobility is more important than specific technology * 00:30:45 – How does “Day 2” operations affect technology choices? * 00:32:30 – Do three-tier architectures still have relevance? * 00:35:00 – Technology change is driving organisational change * 00:38:00 – How can we build in consumption models to technology adoption? * 00:39:00 – Conclusions

Related Podcasts & Blogs

* #162 – The Evolving Role of Storage Administration* #142 – Storage, Automation & DevOps* #104 – Creating a Data Management Strategy* Storage Management and AIOps* Storage Management and DevOps

The role of the storage administrator has changed over the years and is in many ways influenced by the three “V”s typically used to describe AI. Admins have to deal with increased data Volumes, with the Velocity of delivering resources and with a wide Variety of application profiles. This week, Chris talks to Clay Ryder (Senior Director of Enterprise Marketing) and Frank Shepherd (Principal Systems Engineer) from Tintri to gain their insights on how technology and process is changing storage administration to meet the Three Vs challenge.

Exponential data growth has always been an issue in IT. As the cost of storage drops, the volume of data increases, making the work of the storage admin further disconnected from hardware management. Server virtualisation abstracted much of the provision work of storage administration but hasn’t changed fundamentals like workload placement.

Going forward, we need to see technology taking control of the more mundane and repetitive tasks of managing storage infrastructure. AIOps is building that process into storage platforms. At that point, the admin becomes the strategist and architect of business requirements.

You can learn more about Tintri and the VMStore platform at https://tintri.com/. Follow Tintri blogs at https://tintri.com/blog/.

Elapsed Time: 00:43:16

Timeline

* 00:00:00 – Intros* 00:01:00 – Tintri is now part of DDN* 00:01:55 – What is the history of storage administration?* 00:04:55 – Virtualisation didn’t change the construct of a disk* 00:05:40 – Applications & hardware used to be tightly coupled* 00:07:20 – Admins used to spend lots of time managing hardware * 00:09:00 – Storage is a performance arms race* 00:10:50 – Exponential growth doesn’t mean exponential  numbers of staff* 00:13:00 – AIOps will be essential – augmentation like car navigation systems* 00:18:50 – Storage needs to self-adapt to remove administrator intervention* 00:19:30 – Storage needs to adapt to application & business specific needs* 00:22:00 – Virtualisation still depends on underlying storage hardware characteristics* 00:27:30 – (Storage) infrastructure is still needed across the industry * 00:29:30 – We need good analytics from storage platforms * 00:33:50 – Some tasks may be dispersed to non-storage skilled teams * 00:35:20 – What will the future storage administrator skill set look like? * 00:36:20 – The Human aspect moves towards supporting the needs of the business * 00:38:30 – Metcalfe’s Law predicts exponential increase in complexity * 00:42:00 – Wrap Up

Related Podcasts & Blogs

* Storage Management and AIOps* Storage Management and DevOps* #142 – Storage, Automation and DevOps* #114 – CIO storage & Data Challenges

Copyright (c) 2016-2020 Storage Unpacked. No reproduction or re-use without permission. Podcast episode #4m21.

This week, Chris and Martin look at the dual actuator drive technology developed by Seagate, also known as MACH.2. This conversation with Colin Presly and Tim Walker digs down into the detail of the hardware implementation and the operational aspects of deploying a hard drive that acts as two physically independent units.

Dual actuator technology was developed to overcome the performance limitations of traditional hard drives. Over the years, drive capacities have increased almost exponentially, while performance has followed a linear track. MACH.2 divides the read/write heads of a traditional drive into two separate components, each of which can access their respective disk platters independently.

Dual actuator technology introduces some operational challenges for customers. Although the drives act as two unique LUNs on a SAS bus, the physical connectivity requires some thinking around data resiliency and performance optimisation. Unless both actuators are fully utilised, customers won’t see the 2x benefit in performance boost from the technology.

Seagate has worked closely with partners like Microsoft to ensure MACH.2 delivers to the needs of hyperscalers. SDS vendors have also taken an interest, so we can expect to see the technology in a wide range of use cases.

You can find more details on MACH.2 at the Seagate landing page (https://www.seagate.com/gb/en/solutions/mach-2-multi-actuator-hard-drive/) and in this post with Microsoft – https://blog.seagate.com/enterprises/microsoft-nearly-doubles-iops-seagate-exos-mach2-dual-actuator/

Elapsed Time: 00:42:18

Timeline

* 00:00:00 – Intros * 00:01:10 – What is the Seagate Dual Actuator drive? (Mach2) * 00:04:00 – Flash is encroaching the high-end of drives * 00:05:15 – Customers want deterministic performance * 00:06:45 – What is an “actuator”? * 00:10:00 – Multi-actuator can introduce vibration * 00:11:50 – Drive heads float above the disk in Angstroms * 00:12:50 – Do drives still read/write at the cylinder level? * 00:15:10 – Two actuators doubles performance * 00:16:40 – Both parts of the drive are effectively independent * 00:19:00 – Multiple features ensure the drive acts as two entities * 00:21:20 – Deployment scenarios – desktop, hyper-scaler and hardware RAID * 00:22:30 – How is AFR affected by running two drives? * 00:25:00 – Drive reliability is not reduced by having dual actuators * 00:26:10 – Drives have LUN-based and device-based characteristics * 00:27:55 – Microsoft asked for independent devices in Azure * 00:29:10 – SDS vendors are interested in dual actuator drives * 00:30:25 – How will SATA and NVMe interfaces be managed? * 00:35:00 – LBA performance is non-uniform with new drive types * 00:37:00 – Hyperscalers want to eliminate tail latencies  * 00:38:30 – Applications – CDNs, tiering, hyperscale * 00:39:25 – Will we see multi-actuator drives?

Related Podcasts & Blogs

* #55 – Storage for Hyperscalers* #141 – Building Storage Systems of the Future* Unreliable Disks for Better Scale-Out Storage*