Vianney Rancurel retweeted
Super impressive. Congrats!
And maybe suggests that during RSI, the automated AI researchers will be less bottlenecked by compute than we might naively think.
Frontier pretraining is said to be a big-lab-only game. We don’t have 100k chips yet, so there’s only one way: algorithmic efficiency. Our new recipe matches DeepSeek V4 Pro’s pretrain using 50x less compute – that’s roughly half the FLOPs used for GPT3, or ~$0.5M on GB200.
magic.dev/blog/pretraining
Amazed by what our team has achieved ! Next stop: RL 🚀
Frontier pretraining is said to be a big-lab-only game. We don’t have 100k chips yet, so there’s only one way: algorithmic efficiency. Our new recipe matches DeepSeek V4 Pro’s pretrain using 50x less compute – that’s roughly half the FLOPs used for GPT3, or ~$0.5M on GB200.
magic.dev/blog/pretraining
Vianney Rancurel retweeted
We're on tour with @Veeam, sharing how you can bulletproof your #data backup plan. It's not too late to sign up! #tech eventbrite.com/e/scality-at-…
Vianney Rancurel retweeted
By 2030, an estimated 50,000 satellites will zoom through our night sky. That's a lot of #aerospace #data. See how Scality is helping manage and store it. solved.scality.com/solved/sc…
Vianney Rancurel retweeted
A major U.S. bank needed a storage solution for its 80-petabyte #data lake deployment – one that offered secondary #disaster recovery and fast recovery point objective. Learn how Scality #RING stood up to the challenge here: scality.com/press-releases/s… #disasterrecovery
Vianney Rancurel retweeted
In #financial services? Meet regulatory compliance standards with #security features and capabilities for auditing, #DataRetention, legal holds, and data immutability (WORM) with #ScalityRING. Get the report from @CohassetAssoc. #finance #dataprotection scality.com/landing/comply-c…
Vianney Rancurel retweeted
2nd Round of Speakers Announced for SODACON.
This week we continue bringing you another awesome lineup of speakers that will be a part of SODACON.
See more speakers on our event page and register today to join us on July 13-14: bit.ly/3g9yg9k
Vianney Rancurel retweeted
After designing #CSI for block and file #storage volume integration in #Kubernetes, the community now works on #COSI, the #Container #ObjectStorage Interface. In this article, we outline how #object storage provisioning will be integrated in #K8s #CloudNative applications.
Beyond Block and File: COSI Enables Object Storage in Kubernetes bit.ly/3nel71I @eikke #DariaGurova @CloudNativeFdn #Sponsored @scality @KubeCon_ #CloudNativeCon #Kubernetes
Massive storage for AI and Analytics
ICYMI @Utilizing_AI Podcast: AI and Analytics Are Driving a New Kind of Storage with Brad King of Scality @SFoskett @ChrisGrundemann @Scality @Baslking #AI bit.ly/3ruQSVO
Vianney Rancurel retweeted
MARCH 10: Join us for #CFD10 @TechFieldDay @GestaltIT! Curious about what we’re presenting? Hear it directly from one of our presenters - @pspeciale, CPO. #datasovereignty #digitalsovereignty #cloud #datamanagement
Vianney Rancurel retweeted
Read Andrea Fabrizi's post @HPE_AI /@HPE_GreenLake about the new #datamanagement paradigm & optimizing data accessibility for #AI applications. Data is always accessible, and it doesn't ever age. #ml #datalake #artificialintelligence #machinelearning
community.hpe.com/t5/Advanta…
Vianney Rancurel retweeted
If you're using the #MetalK8s #Kubernetes distribution we build as an #opensource project at @Scality, we'd love to hear about your use-case at github.com/scality/metalk8s/… 🙏 #Community input is much appreciated 🤗
Vianney Rancurel retweeted
Andea Fabrizi @HPE_GreenLake shares:"The new #datamanagement paradigm is based upon optimizing data accessibility for #AI applications...Data is always accessible, and it doesn't ever age." #artificialintelligence #machinelearning #dataanalytics #datalakes community.hpe.com/t5/Advanta…
Vianney Rancurel retweeted
@pspeciale & @eikke speak about cloud native data management made easy using Zenko at #sodacon2020.
Vianney Rancurel retweeted
We're very excited to be at @sodafoundation #sodacon2020! Hope you can join us and hear the talk by @pspeciale and @eikke.
The next talk on day 1 of #sodacon2020 is Cloud Native Data Management using Zenko by Paul Speciale & Nicolas Trangez @eikke . Learn how it can provide context for the multi-cloud use-cases that SODA will address.
More: sodafoundation.io/events/sod…
#OneDataFramework
Vianney Rancurel retweeted
One of our presentations from #CFD9. This one by @_rahulreddy, our Director of Engineering. Enjoy!
ICYMI @Scality #CFD9 Video: Building Eventually Consistent Views from AWS S3 with Scality tfd.bz/3n91I1a
Vianney Rancurel retweeted
As you must have realized by now, we use #Kubernetes quite a bit at @Scality. Here's a blog-post discussing some experiences and insights, based on my #CFD9 talk. Looking forward to hear/read your thoughts!
New blog post out!
Scality's journey through Kubernetes now spans the better part of four years.
scality.com/solved/our-exper…
#kubernetes #nodejs #docker #zenko #scality #ring
Vianney Rancurel retweeted
1 Tb/s - the equivalent of 50 HD movie every second ! and 10x cheaper! that's what Scality delivers in Microsoft Azure. When we talk about performance and cost reduction we mean it ! Congrats to Scality R&D team!!!!
Vianney Rancurel retweeted
Scality’s SOFS in Azure makes Blobs blindingly fast dlvr.it/RhgXzg

























