Store
Micron Is Fashionably Late To The HBM Party, But Not Too Late
Here is what memory bandwidth and a certain amount of capacity is worth in the GenAI revolution. …
Memory
Store
We Can’t Get Enough HBM, Or Stack It Up High Enough
There are lots of ways that we might build out the memory capacity and memory bandwidth of compute engines to drive AI and HPC workloads better than we have been able to do thus far. …
Compute
How MCR Memory Can More Than Double HPC And AI Performance
Intel recently demonstrated a new type of DIMM memory technology called Multiplexer Combined Rank (MCR), also referred to as MRDIMMs, that provides up to 2.3X better performance for HPC workloads and up to a 2X better on AI inference workloads in comparison to 5th Gen Intel Xeon processors, based on internal Intel analysis. …
Cloud
Kubernetes Clusters Have Massive Overprovisioning Of Compute And Memory
In the ten years since Google released Kubernetes to the open source community, it has become the dominant platform for orchestrating and managing software containers and microservices, along the way muscling out competitors like Docker Swarm and Mesosphere. …
Connect
Photonics To Make Celestial HBM3 Memory Fabric
There is no shortage of silicon photonics technologies under development, and every few months it seems like another startup crops up promising massive bandwidth, over longer distances, while using less power than copper interconnects. …
Connect
Rest In Pieces: Servers And CXL
If you had to rank the level of hype around specific datacenter technologies, the top thing these days would be, without question, generative AI, probably followed by AI training and inference of all kinds and mixed precision computing in general. …
Compute
Compute Is Easy, Memory Is Harder And Harder
What good is a floating point operation embodied in a vector or matrix unit if you can’t get data into fast enough to actually use the compute engine to process it in some fashion in a clock cycle? …
Store
Just How Bad Is CXL Memory Latency?
Conventional wisdom says that trying to attach system memory to the PCI-Express bus is a bad idea if you care at all about latency. …
Store
Riding CXL Memory Up In A Down Economy
The jittery economy hasn’t been kind to most semiconductor companies, even those like AMD and Nvidia that are growing in the datacenter. …
Connect
Mashing Up CXL And Gen-Z For Shared Disaggregated Memory
If you are impatient for not just memory pooling powered by the CXL protocol, but the much more difficult task of memory sharing by servers attached to giant blocks of external memory, you are not alone. …