Technology:Unpacked

In this podcast, originally recorded on 30 October 2017, Chris and Gavin follow up on the subject of disaggregated storage with Zivan Ori, CEO at E8 Storage.  E8 is a start-up that markets and sells a storage array that uses a disaggregated architecture.

The conversation covers what disaggregation really means and how E8 focuses on splitting the data and control paths of traditional I/O commands.  With NVMe and RDMA, the E8 architecture allows direct access to metadata on dual controllers within the E8 hardware.  This piece was actually a revelation during the conversation as it explains how shared direct access to NVMe drives can be achieved without each connected host getting in the way of another.

The conversation wraps up with a discussion on typical customers and what applications need this level of high performance and low latency.  The original podcast on disaggregation can be found here – Garbage Collection #005 – Disaggregated Storage.

Note that this podcast is a little longer than our normal shows, however it is packed with some great insights into the use of NVMe versus SAS/SATA drives and much more.

Elapsed time: 00:41:36

Timeline

* 00:00:00 – Introductions

* 00:01:00 – Disaggregated Storage recap

* 00:02:00 – Separating the data and control paths, why this matters

* 00:03:30 – How much better are NVMe drives than SAS/SATA? (Including Zivan’s laptop)

* 00:05:30 – What about the price differential – are NVMe drives more expensive per IO?

* 00:06:53 – Apologies for the background sneeze….

* 00:07:23 – Betting on commodity trends – not taking the DSSD route (or Violin)

* 00:08:20 – Getting into more technical detail

* 00:10:00 – How can you break the dual controller model with just software?

* 00:10:52 – Using RDMA to extend the control and data planes outside the shared array

* 00:11:36 – How important is networking in this architecture?

* 00:12:35 – Fibre Channel doesn’t cut it any more?

* 00:14:06 – Why not build a server solution, rather than a storage solution?

* 00:15:48 – Shared storage still provides an advantage over hyper-converged

* 00:17:00 – Getting into E8 architecture in more detail

* 00:18:52 – Client requirements, drivers, agents and emulated block devices

* 00:20:00 – The secret sauce – splitting the data and control paths for generic I/O commands

* 00:25:20 – and here’s the key – being able to talk to the metadata servers with very low latency

* 00:27:00 – New architectures – driven by RDMA and NVMe

* 00:28:48 – How is data protection within the box implemented?

* 00:32:30 – What does a typical customer look like?

* 00:38:00 – How does faster storage lead to better business advantage?

* 00:39:17 – What’s next?

* 00:40:50 – Wrap up time

Vendors/Products Mentioned In this Podcast

* DSSD, E8 Storage, Pure Storage.

Copyright (c) 2016-2017 Storage Unpacked.  No reproduction or re-use without permission.

In this podcast, originally recorded on 30 October 2017, Chris and Gavin follow up on the subject of disaggregated storage with Zivan Ori, CEO at E8 Storage.  E8 is a start-up that markets and sells a storage array that uses a disaggregated architecture. The conversation covers what disaggregation really means and how E8 focuses on […]

This week LTO-8 was announced, with raw cartridge capacities of up to 12.8TB.  In this podcast, recorded on 23rd October 2017, Chris and Martin discuss the evolution of LTO from a lowly 100GB to todays media with 120x the capacity.

Tape is changing and becoming more of an archive medium, as backup has increasingly moved to disk.  This begs the question, should standard formats like LTFS be more common?  How should media be recycled and what other ways are there for getting data onto tape media?

For a bit of nostalgia, there’s mention of the Microdrive from Sinclair and the once ubiquitous Jazz and ZIP drives.

Don’t forget to rate us on iTunes, it only takes a moment!

Elapsed Time: 00:27:46

Timeline

* 00:00:00 – Introductions

* 00:01:10 – LTO-8 is released – LTO speeds and feeds (link)

* 00:04:00 – Positioning against the competition – 3592 and T10000, LTO competes

* 00:06:40 – The changing position of tape

* 00:08:00 – Backup has changed to re-provisioning, with restore only for the data

* 00:10:10 – What happened to LTFS? Still used to ship content around (link)

* 00:11:45 – Spectra Logic Black Pearl – a gateway to getting data onto tape (link)

* 00:13:20 – Tape as the perfect medium for archive

* 00:15:00 – LTO-8-M 50% extra capacity (link)

* 00:17:00 – Jaz and ZIP Drives! (link)

* 00:18:10 – The ZX Spectrum Microdrive (link)

* 00:20:30 – Archive and compliance for tape

* 00:22:50 – Refresh and maintaining an archive

* 00:24:30 – The hundred year archive

* 00:25:00 – The effect of GDPR on archiving and data deletion with tape

Companies/Products Mentioned in this Podcast

Iomega, HPE, Quantum, IBM, Sinclair Research, Spectra Logic, Black Pearl

Copyright (c) 2016-2017 Storage Unpacked.  No reproduction or re-use without permission.

This week LTO-8 was announced, with raw cartridge capacities of up to 12.8TB.  In this podcast, recorded on 23rd October 2017, Chris and Martin discuss the evolution of LTO from a lowly 100GB to todays media with 120x the capacity.

Tape is changing and becoming more of an archive medium, as backup has increasingly moved to disk.  This begs the question, should standard formats like LTFS be more common?  How should media be recycled and what other ways are there for getting data onto tape media?

For a bit of nostalgia, there’s mention of the Microdrive from Sinclair and the once ubiquitous Jazz and ZIP drives.

Don’t forget to rate us on iTunes, it only takes a moment!

Elapsed Time: 00:27:46

Timeline

* 00:00:00 – Introductions

* 00:01:10 – LTO-8 is released – LTO speeds and feeds (link)

* 00:04:00 – Positioning against the competition – 3592 and T10000, LTO competes

* 00:06:40 – The changing position of tape

* 00:08:00 – Backup has changed to re-provisioning, with restore only for the data

* 00:10:10 – What happened to LTFS? Still used to ship content around (link)

* 00:11:45 – Spectra Logic Black Pearl – a gateway to getting data onto tape (link)

* 00:13:20 – Tape as the perfect medium for archive

* 00:15:00 – LTO-8-M 50% extra capacity (link)

* 00:17:00 – Jaz and ZIP Drives! (link)

* 00:18:10 – The ZX Spectrum Microdrive (link)

* 00:20:30 – Archive and compliance for tape

* 00:22:50 – Refresh and maintaining an archive

* 00:24:30 – The hundred year archive

* 00:25:00 – The effect of GDPR on archiving and data deletion with tape

Companies/Products Mentioned in this Podcast

Iomega, HPE, Quantum, IBM, Sinclair Research, Spectra Logic, Black Pearl

Copyright (c) 2016-2017 Storage Unpacked.  No reproduction or re-use without permission.

This week LTO-8 was announced, with raw cartridge capacities of up to 12.8TB.  In this podcast, recorded on 23rd October 2017, Chris and Martin discuss the evolution of LTO from a lowly 100GB to todays media with 120x the capacity. Tape is changing and becoming more of an archive medium, as backup has increasingly moved […]

In this podcast, recorded on 9 October 2017, Chris and Martin chat to Kieran Harty, CTO and co-founder at Tintri.  The discussion covers the requirements for private cloud and in particular how storage should be deployed in private cloud environments.  Kieran highlights how Tintri has adapted their all-flash and hybrid arrays to meet the needs of organisations looking to transform to private cloud operations.

During the podcast Kieran mentions Tintri Scale-out.  Details can be found here.  Chris mentions the HPE 3PAR paper on private cloud storage, which can be found here.

Elapsed Time: 00:32:15

Timeline

* 00:00:00 – Intro

* 00:01:00 – Explaining Private Cloud (Storage) with the NIST definition

* 00:03:30 – Performance isn’t part of the NIST definition – but still important

* 00:04:00 – Why is delivering private cloud storage hard?

* 00:05:00 – Looking at VMs rather than LUNs and volumes – the right abstraction

* 00:07:30 – Applying policies to virtual machines

* 00:08:30 – Self service and automation – a critical function

* 00:11:50 – VMAX Cloud Edition and the HPE Private Cloud Storage

* 00:12:50 – Quotas, at the storage and at the hypervisor management level

* 00:13:30 – Multi-tenancy and the noisy neighbour, QoS at the VM level

* 00:16:00 – Federation, moving workload between multiple discrete appliances

* 00:20:35 – Tintri Snapshot feature

* 00:24:30 – DevOps, Slack and Chris’s inability to remember we’re not talking to computers

* 00:27:30 – I’m sorry Dave, but I can’t do that

* 00:28:00 – How does Private Cloud storage affect the cost model and TCO?

* 00:30:00 – What about Container support with Tintri?

* 00:31:30 – Wrap Up

Copyright (c) 2016-2017 Storage Unpacked.  No reproduction or re-use without permission.

In this podcast, recorded on 9 October 2017, Chris and Martin chat to Kieran Harty, CTO and co-founder at Tintri.  The discussion covers the requirements for private cloud and in particular how storage should be deployed in private cloud environments.  Kieran highlights how Tintri has adapted their all-flash and hybrid arrays to meet the needs […]

In this podcast, recorded live on 13th October 2017 in London, Chris talks to Violin Systems CEO, Ebrahim Abbasi.  Violin Systems is the new name for Violin Memory, one of the pioneers of all-flash storage.  Violin Memory entered chapter 11 in December 2016 and the company assets were eventually acquired by the Soros Group.

Violin Systems is back, with no debt, positive cashflow and plans for a strong future.  Chris and Ebrahim discuss the software acquired from the relationship with Falconstor and taking a more software-focused approach to product development.  We can expect scale-out, tiering and SDS-based solutions in the future.  The discussion finishes up with a conversation about how the company will work to become credible and compete in an increasingly competitive all-flash market.

Elapsed Time: 00:19:15

Copyright (c) 2016-2017 Storage Unpacked.  No reproduction or re-use without permission.

In this podcast, recorded live on 13th October 2017 in London, Chris talks to Violin Systems CEO, Ebrahim Abbasi.  Violin Systems is the new name for Violin Memory, one of the pioneers of all-flash storage.  Violin Memory entered chapter 11 in December 2016 and the company assets were eventually acquired by the Soros Group. Violin […]

In this podcast, recorded on 2nd September 2017, the guys chat to Andy Hardy, Regional Vice President at Nasuni.  Nasuni has developed a global file platform that uses cloud storage as the “backing store”.  In this podcast, the guys chat to Andy about the benefits of a global file system that requires very little on-premises storage and the components it replaces, including backup and DR.  The conversation also touches on where the market is headed, who is having success and the continual divergence between IOPS for block and capacity storage for file/object.

Elapsed Time: 00:29:09

Timeline

* 00:00:00 – Introductions

* 00:01:00 – Nasuni – a cloud-native file system, file backed by object storage

* 00:02:00 – Cloud File Storage – replacing onsite backup, storage and DR

* 00:03:00 – NetApp, Isilon killer!

* 00:04:00 – Obfuscating the cloud provider and choosing the right platform

* 00:05:00 – Data sovereignty and compliance – managing data for a global enterprise

* 00:06:00 – Mixing and matching object storage providers – is this a commodity game already?

* 00:08:00 – Storage ratios – 20% block, 80% unstructured – but block sees the money

* 00:10:00 – Solving the latency problem with local caching appliances – not specific data placement

* 00:12:00 – What are the use cases?  Can this work for analytics? Four use cases – Global NAS, Collaboration, Active Archive and Analytics

* 00:13:50 – IBM Newton?  Surely IBM Watson, Andy…and Donald Trump semi-naked….

* 00:15:00 – Partner relationships – who wants to work with global NAS providers?

* 00:15:20 – Microsoft being Microsoft and the success of Azure

* 00:17:20 – What about the rest of the storage industry?

* 00:17:40 – The changing economics of deploying storage

* 00:19:30 – The shrinking and increasingly competitive block storage market

* 00:20:30 – As (unstructured) data moves to the cloud, on-prem focuses on IOPS spending

* 00:20:55 – How do you deal with failure? (Not a personal question, Andy)

* 00:24:00 – What about getting access to my data directly in the multi-cloud?

* 00:27:00 – What is up and coming for Nasuni?  Integration and APIs

* 00:28:00 – Wrap up!

Vendors Mentioned In This Podcast

These vendors, sometimes briefly, are all mentioned in this podcast episode.

Nasuni, Azure Blob, AWS S3, Nirvanix, IBM Cloud Object Storage, Cleversafe, EMC ECS, Cloudian, Dell Storage, Compellent, Nutanix, Tintri, Tegile, Nimble Storage, Pure Storage, 3PAR, LeftHand, Equallogic, Simplivity

Copyright (c) 2016-2017 Storage Unpacked.  No reproduction or re-use without permission.