mcrutcher

7 ms·

1. 1.▲
   by mcrutcher 1y ago

   Also, though nobody has put the work in yet, the GH200 and GB200 (the NVIDIA "superchips" support exposing their full LPDDR5X and HBM3 as UVM (unified virtual memory) with much more memory bandwidth between LPDDR5X and HBM3 than a
2. 2.▲
   by mcrutcher 1y ago

   MoE models are pretty poorly named since all the "experts" are "the same". They're probably better described as "sparse activation" models. MoE implies some sort of "heterogenous experts" that a
3. 3.▲
   by mcrutcher 5y ago

   If they're almost at cost parity now it seems like a sure bet that you'd have to have a very abnormal use case to consider a gas vehicle 5 years from now.
4. 4.▲
   by mcrutcher 6y ago

   This seems highly relevant as to what is actually going on: http://www.johncostella.com/magic/
5. 5.▲
   by mcrutcher 6y ago

   I was the PM for the software side of Greenplum DB for a while. I also don't know of an earlier example of a separate column oriented storage option on top of a Postgres base. Vertica and ParAccel (which became Redshift) both also star
6. 6.▲
   by mcrutcher 6y ago

   This has existed for a very long time as a data modeling strategy (most commonly, a "type 2 dimension") and is the way that all MVCC databases work under the covers. You don't need a special database to do this, just add anot