NVIDIA Acquires SwiftStack, an Object Storage Company
by Anton Shilov on March 10, 2020 1:00 PM EST- Posted in
- Datacenter
- GPUs
- NVIDIA
- Deep Learning
- AI
- SwiftStack
NVIDIA and SwiftStack have signed an agreement under which the former will aqcuire the latter. SwiftStack develops object storage software that is used for AI, HPC, and accelerated computing applications. As a part of NVIDIA, SwiftStack will continue supporting its existing programs and develop new ones, primarily with NVIDIA’s hardware in mind.
SwiftStack was founded in November 2011 and raised $23.6 million in total. The company is primarily known for its SwiftStack 7, 1space, ProxyFS, Swift, and the Controller products, but market observers do not think that it was a really successful company. Throughout its history, SwiftStack has worked with many companies, including Ebay, Google, NVIDIA, and Valohai. Furthermore, the company has experience with hardware from Cisco, Dell, HPE, and SuperMicro.
After it is integrated as a part of NVIDIA, SwiftStack will continue to develop, support, and enhance its programs. Meanwhile, the key mission of the team will be to work on NVIDIA’s GPU-powered AI infrastructure and accelerate large scale AI DL pipelines.
In the recent years, NVIDIA has been meticulously collecting technologies and IP for datacenters, with supercomputers in particular being a key aspiration, which emphasizes the focus of the company which has moved beyond just computer graphics in all of its forms. Last year NVIDIA announced a plan to acquire Mellanox, which makes connectivity solutions for servers. The takeover of SwiftStack, an object storage company, is another way to collect assets required for scalable GPU-powered infrastructure for deep learning.
Terms and conditions of the deal were not disclosed, but the two companies expect the transaction to close in the coming weeks.
Related Reading:
- NVIDIA To Acquire Datacenter Networking Firm Mellanox for $6.9 Billion
- NVIDIA Axes GTC Digital Keynote In Favor Of News Releases; Fate Of Major Announcements Unknown
- NVIDIA's 2020 GTC Becomes Latest Coronavirus Event Casualty: Now Digital-Only
- NVIDIA Releases Q4 FY2020 Earnings: A Strong Quarter Closes Out A Weaker Year
Sources: SwiftStack, StorageNewsletter
Facebook
Twitter
RSS
5 Comments
View All Comments
jeremyshaw - Tuesday, March 10, 2020 - link
I'm somewhat struggling to figure out what SwiftStack does that is special.It does seem to be a software(?) abstraction layer for massive and varied storage systems?
Is/Are their key advantages scalability? Flexability? Performance across distant supercomputing clusters/centers?
What is their advantage in Nvidia's portfolio? Is Nvidia doing this as a defensive maneuver, like Mellanox? Is there some sort of special IP involved? Is the real aquisition the talent?
Is this a ramp up for GTC news releases and investor call? Is the aquisition cost (not disclosed? not finished?) small enough to not warrent regulator approval?
Are their offices close enough to merge, or are we looking at an Icera-like situation, where the other team is still largely doing their own thing?
olafgarten - Tuesday, March 10, 2020 - link
From what I can tell, they have two products, one named Storage which is an on-prem object storage program, sort of like MinIO, which lets you also access it as an NTFS/SMB store.Their second product is 1space, and seems to be a multi-cloud object storage, in that data can be stored in many different places but accessed centrally. I can't see why anyone would need to use this though.
PaulHoule - Sunday, March 15, 2020 - link
Storage companies get acquired for no reason all the time, probably because they fail and can get picked up cheap.The diagram that shows Swiftstack used for ML training data looks delusional. If you are tagging images, for instance, you will probably put the images in an object store. The tags are really small in comparison and are a better fit for a database, particularly if you want to spool some subset of them to the trainer in a hurry.
Lakados - Monday, March 16, 2020 - link
They own the IP for a couple of file storage systems and file formats that improve throughput speeds related to AI specific tasks that assist in reducing training time on images in particular, but could be expanded on with the right research.spaces - Monday, March 16, 2020 - link
Worth going through this article. Thanks for providing full detailed information in this post https://happywheelspaces.com