December 2021 saw the return of an in person user group event as part of the annual CIUK conference in Manchester. Whilst it wasn’t the busiest SSUG we’ve run, it was great to see colleagues back and talking about Scale. The event was of course held in-line with government guidelines and all attendees were asked to wear masks throughout CIUK – and for the most part, people did do this!
The sessions kicked off with Indulis Bernsteins (IBM), an SSUG regular speaker, presenting on what’s new in 5.1.3. Whilst we’ve run a few “What’s new” sessions digitally, actually being in the room and having people be able to ask questions directly was a great return. (Personally, I’ve talked and presented at a few digital/online events over the past year or two, and the lack of any kind of response from the audience is what I’ve found most difficult, so it was great to actually see people respond in person!)
Steve Hindmarsh, Francis Crick Institute then presented an update on Scale at the Crick, as well as talking about the older systems, Steve also gave us a little sneak peak at the new storage going in in 2022.
Daniel Kidger followed on, Daniel is also a familiar speaker at SSUG, but now is at HPE. Since moving there, he’s been pulled in to working with the ECE storage systems now shipped by HPE and gave an update on the licensing model, servers as well as some detail on scaling and performance of the systems. … I think this is the first production system ECE talk we’ve had so far at SSUG!
Finally, Indulis returned to the stage giving an update on Spectrum Discover and use cases with Spectrum Scale.
I’d like to extend a personal thanks to all the speakers at CIUK for agreeing to even speak in person, as well as to all the attendees who made it and joined us for the event. Actually, not just for the CIUK21 event, but over the past few years – due to a change in role, I’m standing back from running the user group. We even got a volunteer at CIUK to agree taking on running events in the UK! A big shout out of thanks should go to Ulf Troppens, from IBM who has worked with the user group over the past few years, finding speakers and topics that fit the user group agenda! The past two years has been difficult with no in-person events, so actually getting to do one as my last running the group was a great opportunity… I’m sure I’ll still see many of the SSUG colleagues in the future, I’ve met some great people in my time running the group and that has supported some interesting projects as well as giving some fantastic conversations. I’m sure the user group will continue to thrive, and with restrictions on events changing, maybe we’ll see some more in-person opportunities and get to meet in person again!
Due to the ongoing COVID-19 outbreak, this in-person event is cancelled. We hope to bring a digital version!
The new user day is something we’ve tried to squeeze around the main agenda in the past, this year is running as a completely separate day prior to the main event.
Due to the ongoing COVID-19 outbreak, this in-person event is cancelled. We hope to bring a digital version!
This year’s Spectrum Scale user group event has a fantastic agenda of Spectrum Scale experts and will take place over 2-days, again at IBM’s client centre overlooking the River Thames in London. 2020 sees 10 years of the user group in the UK! The event will provide you with the opportunity to hear the latest Spectrum Scale updates from filesystem experts as well as hear from other users on their experiences. Similar to previous years, we’re also holding smaller interactive workshops to allow for more detailed discussion.
The UK/World Wide 2019 Spectrum Scale user group is planned to take place on 8th and 9th May, once again overlooking the River Thames at IBM’s South Bank Client Centre.
The draft agenda for the event is posted below and has a mix of IBM, customer and sponsor talks. This year we have a particular focus on AI with Spectrum Scale and it promises to be another great two days.
Thanks to our sponsors, we’re able to run an evening dinner/networking event. We’ll provide more details on this on the day, but make sure you plan to attend our evening event!
Agenda
Day 1: 8th May:
Start
End
Title
Speaker
60
09:00
10:00
ARRIVALS & REGISTRATION
15
10:00
10:15
Opening & welcome
Simon Thompson (Chair) / Ted Hoover (IBM)
25
10:15
10:40
Spectrum Scale: The past, the present, the future
Wayne Sawdon (IBM)
25
10:40
11:05
Accelerating AI workloads with IBM Spectrum Scale
Ted Hoover (IBM)
20
11:05
11:25
Sponsor Talk [NVIDIA]: Doing large scale AI faster – scale innovation with multi-node-multi-GPU computing and a scaling data pipeline
Miguel Martínez (NVIDIA)
25
11:25
11:50
AI ecosystem and solutions with IBM Spectrum Scale
Piyush Chaudhary (IBM)
10
11:50
12:00
Meet the devs
Ulf Troppens (IBM)
60
12:00
13:00
LUNCH
30
13:00
13:30
Spectrum Scale Update
Puneet Chaudhary
15
13:30
13:45
ESS Update
Puneet Chaudhary
20
13:45
14:05
Customer Talk - Why Spectrum Scale, a Customer View
Mark Vilensky (Weizmann Institute of Science)
20
14:05
14:25
Sponsor Talk [DDN]: "Optimizing Storage Stacks for AI"
Sven Oehme
20
14:25
14:45
Sponsor Talk [Lenovo] - "Storage, Fast and Slow - Lenovo updates from both ends of the storage hierarchy"
Michael Hennecke
30
14:45
15:15
BREAK
15
15:15
15:30
Support Update
Kai Korkow (IBM)
20
15:30
15:50
New Storage Options for Spectrum Scale
Luis Bolinches (IBM)
20
15:50
16:10
Customer Talk - Why Spectrum Scale, a Customer View
Patrick Dekkers (Amsterdam UMC Locactie VUmc)
20
16:10
16:30
Sponsor Talk [e8 Storage]: E8 Shared NVMe for High Performance Spectrum Scale Clusters
Stuart Campbell (e8 Storage) / Peter Childs (QMUL)
During the afternoon, we will also be running a breakout session “Getting Started with Spectrum Scale”. This is intended for newer users of Spectrum Scale and will be run subject to demand.
Day 2: 9th May:
Start
End
Topic
Speaker
30
08:30
09:00
ARRIVALS
30
08:30
09:00
Spectrum Scale Licensing
Carl Zetie (IBM)
60
09:00
10:00
AI at Scale at the example of Autonomous Driving (Beginner)
IBM Aspera Sync for Spectrum Scale
Nils Haustein (IBM)
Cloud Update
Spectrum Scale Support for Container
Harald Seipp (IBM)
Towards Proactive Service with Call Home
Spectrum Scale Reliability, Availability und Serviceability
Mathias Dietz (IBM)
60
10:00
11:00
Spectrum Scale installation and upgrade with Install Toolkit
Kedar Karmarkar (IBM)
Tiering to tape with Spectrum Archive
Khanh Ngo (IBM)
Spectrum Scale Watch Folder
Lessons from the field
Olaf Weiser (IBM)
Kai Korkow (IBM) / Luis Bolinches (IBM)
60
11:00
12:00
Overview: Spectrum Scale support for NFS and SMB
Spectrum Scale Best Practices for Snapshots
Ulf Troppens (IBM)
Kedar Karmarkar (IBM)
Spectrum Scale Policy Engine: Introduction and Use Cases
Nils Haustein (IBM)
Spectrum Scale Network Flow
John Lewars (IBM)
60
12:00
13:00
LUNCH
30
13:00
13:30
Lightning Talks - My favourite/worst Spectrum Scale feature
All - including attendees
30
13:30
14:00
Memory Consumption in Spectrum Scale
John Lewars (IBM)
30
14:00
14:30
BREAK
20
14:30
14:50
Customer Talk - Migrating Spectrum Scale Filesystems with minimal downtime
Luke Sudbery (University of Birmingham)
20
14:50
15:10
Sponsor Talk [OCF]: Atempo - Large Filesystem Migration & Backup
Thomas Ahrens (Atempo)
20
15:10
15:30
Customer Talk - Experience with Spectrum Scale AFM
Alvise Dorigo (PSI)
20
15:30
15:50
High Performance Tier
Wayne Sawdon (IBM)
10
15:50
16:00
CLOSE
Simon Thompson (Chair)
This event is sponsored by IBM, OCF, e8 Storage, Lenovo, DDN Storage, Mellanox Technologies and NVIDIA.