HPC Guru (Twitter)
.@Broadcom says its Jericho3-AI #Ethernet is better than @NVIDIA’s #Infiniband by ~10% on NCCL performance
https://t.co/Z6zCsyXke1
#AI #Network via @ServeTheHome
.@Broadcom says its Jericho3-AI #Ethernet is better than @NVIDIA’s #Infiniband by ~10% on NCCL performance
https://t.co/Z6zCsyXke1
#AI #Network via @ServeTheHome
ServeTheHome
Broadcom Jericho3-AI Ethernet Switch Launched Saying NVIDIA Infiniband is Bad for AI
In launching the new Broadcom Jericho3-AI Ethernet switch chip, Broadcom is taking an overt swipe at NVIDIA's Infiniband for AI
HPC Guru (Twitter)
The new era needs a new #network: Industry Heavyweights Form Ultra #Ethernet Consortium for #HPC and #AI
Founding members include AMD, Arista, Broadcom, Cisco, Eviden (Atos), HPE, Intel, Meta and Microsoft
https://t.co/BPGZWNeQkW
https://t.co/V4cwgGngDo
via @insideHPC https://t.co/rKVPGrb9fc
The new era needs a new #network: Industry Heavyweights Form Ultra #Ethernet Consortium for #HPC and #AI
Founding members include AMD, Arista, Broadcom, Cisco, Eviden (Atos), HPE, Intel, Meta and Microsoft
https://t.co/BPGZWNeQkW
https://t.co/V4cwgGngDo
via @insideHPC https://t.co/rKVPGrb9fc
HPC Guru (Twitter)
.@Azure #Boost offloads virtualization processes traditionally performed by the hypervisor and host OS, such as networking, storage, and host management, onto purpose-built hardware and software
MANA smartNICs to counter AWS Nitro?
https://t.co/FeKBWzqZup
#Cloud #Network https://t.co/uAlSWmla56
.@Azure #Boost offloads virtualization processes traditionally performed by the hypervisor and host OS, such as networking, storage, and host management, onto purpose-built hardware and software
MANA smartNICs to counter AWS Nitro?
https://t.co/FeKBWzqZup
#Cloud #Network https://t.co/uAlSWmla56
HPC Guru (Twitter)
#SciNet Brings #IPv6 to the Blue Bear and #SC23
“The biggest challenge is getting people to see the value in adopting IPv6, and the capabilities it can bring”
https://t.co/ogNTMHYQgS
#HPC #AI #Network via @Supercomputing
#SciNet Brings #IPv6 to the Blue Bear and #SC23
“The biggest challenge is getting people to see the value in adopting IPv6, and the capabilities it can bring”
https://t.co/ogNTMHYQgS
#HPC #AI #Network via @Supercomputing
SC23
SCinet Brings IPv6 to the Blue Bear and SC23 • SC23
SCinet's work at SC23 will allow people to get hands-on experience with IPv6 and take that knowledge back to their home institutions.
HPC Guru (Twitter)
Why @Meta is determined to make #Ethernet work for #AI
@Meta is one of the founding companies behind the @ultraethernet Consortium, a group of companies -according to @TDaytonPM - united against a common enemy - #Infiniband
nextplatform.com/2023/09/26/…
#HPC #Network
Why @Meta is determined to make #Ethernet work for #AI
@Meta is one of the founding companies behind the @ultraethernet Consortium, a group of companies -according to @TDaytonPM - united against a common enemy - #Infiniband
nextplatform.com/2023/09/26/…
#HPC #Network
HPC Guru (Twitter)
Analysis by @dylan522p on @nvidia multi-pronged strategy to remain the leader for #AI hardware
Nvidia’s plan are to transcend the comparisons with traditional competitors like Intel and AMD and ascend to the rank of tech giants
semianalysis.com/p/nvidias-p…
#GPU #HPC #network
Analysis by @dylan522p on @nvidia multi-pronged strategy to remain the leader for #AI hardware
Nvidia’s plan are to transcend the comparisons with traditional competitors like Intel and AMD and ascend to the rank of tech giants
semianalysis.com/p/nvidias-p…
#GPU #HPC #network
X (formerly Twitter)
HPC Guru on X
Analysis by @dylan522p on @nvidia multi-pronged strategy to remain the leader for #AI hardware
Nvidia’s plan are to transcend the comparisons with traditional competitors like Intel and AMD and ascend to the rank of tech giants
https://t.co/jM9QRYjLVV
…
Nvidia’s plan are to transcend the comparisons with traditional competitors like Intel and AMD and ascend to the rank of tech giants
https://t.co/jM9QRYjLVV
…
HPC Guru (Twitter)
Cornelis Networks announces CN5000 #OmniPath High-Performance #interconnect solution for #AI and #HPC
hpcwire.com/off-the-wire/cor…
#SC23 #Network @CornelisHPC via @HPCwire
Cornelis Networks announces CN5000 #OmniPath High-Performance #interconnect solution for #AI and #HPC
hpcwire.com/off-the-wire/cor…
#SC23 #Network @CornelisHPC via @HPCwire
HPC Guru (Twitter)
Cornelis Networks announces CN5000 #OmniPath High-Performance #interconnect solution for #AI and #HPC
hpcwire.com/off-the-wire/cor…
#SC23 #Network @CornelisHPC via @HPCwire
Cornelis Networks announces CN5000 #OmniPath High-Performance #interconnect solution for #AI and #HPC
hpcwire.com/off-the-wire/cor…
#SC23 #Network @CornelisHPC via @HPCwire
HPC Guru (Twitter)
6.71 terabits per second: This week, Denver will have what may be the fastest #network in the world
And next week? It’ll be gone
coloradosun.com/2023/11/13/f…
#SC23 #SCinet #HPC #AI via @ColoradoSun @ESnet
6.71 terabits per second: This week, Denver will have what may be the fastest #network in the world
And next week? It’ll be gone
coloradosun.com/2023/11/13/f…
#SC23 #SCinet #HPC #AI via @ColoradoSun @ESnet
X (formerly Twitter)
#network - Search / X
See posts about #network on X. See what people are saying and join the conversation.
HPC Guru (Twitter)
Synopsys Launches Industry’s First Complete 1.6T Ethernet IP Solution to Meet High Bandwidth Needs of AI and Hyperscale Data Center Chips
https://news.synopsys.com/2024-02-29-Synopsys-Launches-Industrys-First-Complete-1-6T-Ethernet-IP-Solution-to-Meet-High-Bandwidth-Needs-of-AI-and-Hyperscale-Data-Center-Chips
#HPC #AI #Network #Ethernet
Synopsys Launches Industry’s First Complete 1.6T Ethernet IP Solution to Meet High Bandwidth Needs of AI and Hyperscale Data Center Chips
https://news.synopsys.com/2024-02-29-Synopsys-Launches-Industrys-First-Complete-1-6T-Ethernet-IP-Solution-to-Meet-High-Bandwidth-Needs-of-AI-and-Hyperscale-Data-Center-Chips
#HPC #AI #Network #Ethernet
HPC Guru (Twitter)
With Thor2, @Broadcom wants to replace @nvidia’s ConnectX as the #AI #network adapter
Article also highlights the higher failure rates of optical transceivers
https://www.nextplatform.com/2024/05/28/with-thor-2-broadcom-wants-to-become-the-ai-network-adapter/
#HPC via @TDaytonPM @TheNextPlatform
With Thor2, @Broadcom wants to replace @nvidia’s ConnectX as the #AI #network adapter
Article also highlights the higher failure rates of optical transceivers
https://www.nextplatform.com/2024/05/28/with-thor-2-broadcom-wants-to-become-the-ai-network-adapter/
#HPC via @TDaytonPM @TheNextPlatform
X (formerly Twitter)
HPC Guru (@HPC_Guru) on X
With Thor2, @Broadcom wants to replace @Nvidia’s ConnectX as the #AI #network adapter
Article also highlights the higher failure rates of optical transceivers
https://t.co/qjEbuuILBp
#HPC via @TDaytonPM @TheNextPlatform
Article also highlights the higher failure rates of optical transceivers
https://t.co/qjEbuuILBp
#HPC via @TDaytonPM @TheNextPlatform
HPC Guru (Twitter)
Tencent upgraded its #HPC #network to access idling #GPU capacity offering a 20% improvement in #LLM training
Xingmai network can support a single computing cluster with more than 100,000 GPUs
https://www.scmp.com/tech/big-tech/article/3268901/tencent-boosts-ai-training-efficiency-without-nvidias-most-advanced-chips
#AI #GenAI
Tencent upgraded its #HPC #network to access idling #GPU capacity offering a 20% improvement in #LLM training
Xingmai network can support a single computing cluster with more than 100,000 GPUs
https://www.scmp.com/tech/big-tech/article/3268901/tencent-boosts-ai-training-efficiency-without-nvidias-most-advanced-chips
#AI #GenAI
HPC Guru (Twitter)
.@ServeTheHome reviews the new 51.2T switch based on @MarvellTech's Teralynx10 chip
64 ports 800 GbE
or
512 ports 100 GbE
https://www.servethehome.com/inside-a-marvell-teralynx-10-51-2t-64-port-800gbe-switch/
https://invidious.poast.org/watch?v=Uqx3C3Vn4_g
#HPC #AI #Ethernet #Network @Patrick1Kennedy
.@ServeTheHome reviews the new 51.2T switch based on @MarvellTech's Teralynx10 chip
64 ports 800 GbE
or
512 ports 100 GbE
https://www.servethehome.com/inside-a-marvell-teralynx-10-51-2t-64-port-800gbe-switch/
https://invidious.poast.org/watch?v=Uqx3C3Vn4_g
#HPC #AI #Ethernet #Network @Patrick1Kennedy
ServeTheHome
Inside a Marvell Teralynx 10 51.2T 64-port 800GbE Switch
We dive into a 64-port 800GbE switch. We got to take apart a 51.2T Marvell Teralynx 10 switch before seeing it running traffic
HPC Guru (Twitter)
.@nvidia's #SC24 special address is up on Youtube
https://invidious.poast.org/watch?v=eKzNKxWUeCE
#HPC #AI #GPU #Network
.@nvidia's #SC24 special address is up on Youtube
https://invidious.poast.org/watch?v=eKzNKxWUeCE
#HPC #AI #GPU #Network
HPC Guru (Twitter)
.@EuroHPC_JU is launching a call to build a high-speed, secure #network connecting #European #supercomputers and data centers, creating a hyperconnected and federated #HPC and #QuantumComputing ecosystem
https://eurohpc-ju.europa.eu/eurohpc-ju-launches-call-tender-hyperconnect-european-supercomputers-2024-12-16_en
.@EuroHPC_JU is launching a call to build a high-speed, secure #network connecting #European #supercomputers and data centers, creating a hyperconnected and federated #HPC and #QuantumComputing ecosystem
https://eurohpc-ju.europa.eu/eurohpc-ju-launches-call-tender-hyperconnect-european-supercomputers-2024-12-16_en
HPC Guru (Twitter)
Cornelis Networks CEO Lisa Spelman says their new 400-Gbps CN5000 family of scale-out networking offers the ‘best performance with devastatingly good price-performance,’ and it has bigger plans for the future
https://www.crn.com/news/networking/2025/intel-spin-off-our-infiniband-alternative-for-ai-data-centers-has-a-devastatingly-good-edge
#HPC #AI #Network via @DylanOnChips
Cornelis Networks CEO Lisa Spelman says their new 400-Gbps CN5000 family of scale-out networking offers the ‘best performance with devastatingly good price-performance,’ and it has bigger plans for the future
https://www.crn.com/news/networking/2025/intel-spin-off-our-infiniband-alternative-for-ai-data-centers-has-a-devastatingly-good-edge
#HPC #AI #Network via @DylanOnChips
Crn
Intel Spin-Off: Our InfiniBand Alternative For AI Data Centers Has A ‘Devastatingly Good’ Edge
The CEO of Intel spin-off Cornelis Networks explains to CRN why its new 400-Gbps CN5000 family of high-speed networking products offers the “best performance with devastatingly good price-performance” compared to InfiniBand- and Ethernet-based offerings for…
insideHPC.com (Twitter)
Check out the new article from DriveNets: Re-Engineering Ethernet for AI Fabric
wp.me/p3RLHQ-p7E
@drivenets #Ethernet #Infiniband #AInetwork #network #AI #HPC
Check out the new article from DriveNets: Re-Engineering Ethernet for AI Fabric
wp.me/p3RLHQ-p7E
@drivenets #Ethernet #Infiniband #AInetwork #network #AI #HPC
Inside HPC & AI News | High-Performance Computing & Artificial Intelligence
Re-Engineering Ethernet for AI Fabric
Ethernet wasn’t built with AI in mind. While cost-effective and ubiquitous, its best-effort, packet-based nature creates challenges in AI clusters... But [...]
insideHPC.com (Twitter)
HPE Closes $13.4B Acquisition of Juniper Networks
wp.me/p3RLHQ-p8D
@HPE @JuniperNetworks #AIfactory #AIdatacenter #datacenter #network
HPE Closes $13.4B Acquisition of Juniper Networks
wp.me/p3RLHQ-p8D
@HPE @JuniperNetworks #AIfactory #AIdatacenter #datacenter #network
Inside HPC & AI News | High-Performance Computing & Artificial Intelligence
HPE Closes $13.4B Acquisition of Juniper Networks
HPE today announced the sucessful completion of its acquisition of AI-native networks company Juniper Networks, Inc., initiailly announced in April of last [...]