Timothy Prickett Morgan
Author Archives: Timothy Prickett Morgan
- Home → Author's archive:
Timothy Prickett Morgan
Deep Dive On Google’s Exascale TPUv4 AI Systems
It seems like we have been talking about Google’s TPUv4 machine learning accelerators for a long time, and that is because we have been. …
Deep Dive On Google’s Exascale TPUv4 AI Systems was written by Timothy Prickett Morgan at The Next Platform.
Cores, Clocks, And Caches Cranked With Latest NEC Vector Engines
Whenever a process shrink is available to chip designers, there are several different levers they can pull to make a more powerful compute engine. …
Cores, Clocks, And Caches Cranked With Latest NEC Vector Engines was written by Timothy Prickett Morgan at The Next Platform.
Datacenter Will Be AMD’s Largest – And Most Profitable – Business
Two and a half years into the global coronavirus pandemic we all have upgraded our home IT infrastructure. …
Datacenter Will Be AMD’s Largest – And Most Profitable – Business was written by Timothy Prickett Morgan at The Next Platform.
The Art Of System Design As HPC And AI Applications Diverge
Predicting the future is hard, even with supercomputers. And maybe specifically when you are talking about predicting the future of supercomputers. …
The Art Of System Design As HPC And AI Applications Diverge was written by Timothy Prickett Morgan at The Next Platform.
Different GPU Horses For Different Datacenter Courses
If the semiconductor business teaches us anything, it is that volumes matter more than architecture. …
Different GPU Horses For Different Datacenter Courses was written by Timothy Prickett Morgan at The Next Platform.
The Steady Hand Guiding AMD’s “Prudently Expanding” Datacenter Business
The old AMD – the one before Lisa Su took over – was often brilliant with its instruction set architecture and CPU designs, but sometimes perplexingly careless with its design choices and chip roadmaps. …
The Steady Hand Guiding AMD’s “Prudently Expanding” Datacenter Business was written by Timothy Prickett Morgan at The Next Platform.
Server And Storage Spending Moves The Sticks Out Through 2026
There is a very interesting technical side to IT infrastructure, which we are obviously very keen on exploring here at The Next Platform. …
Server And Storage Spending Moves The Sticks Out Through 2026 was written by Timothy Prickett Morgan at The Next Platform.
Where Amdahl’s Law And Gustafson’s Law Hit the Moore’s Law Wall
After nearly six decades of getting smaller, faster, cooler, and cheaper, transistors are getting more and more expensive with each generation, and one could argue that this, more than any other factor, is going to drive system architecture choices for the foreseeable future. …
Where Amdahl’s Law And Gustafson’s Law Hit the Moore’s Law Wall was written by Timothy Prickett Morgan at The Next Platform.
Intel To Broaden FPGA Lineup And Make Them At Home
Back in 2015, when Intel was flush with cash thanks to a near-monopoly from X86 datacenter compute, it shelled out an incredible $16.7 billion to acquire FPGA maker Altera because a few hyperscalers and cloud builders were monkeying around with offloading whole chunks of CPU compute to FPGAs to create SmartNICs. …
Intel To Broaden FPGA Lineup And Make Them At Home was written by Timothy Prickett Morgan at The Next Platform.
Lenovo Plays The Long Game, Not The Wrong Game, In Systems
It has been nearly four decades since the Chinese Academy of Sciences handed Liu Chuanzhi and Danny Lui $25,000 to help found Legend, originally a maker of TV sets that, in the wake of the success of the IBM PC and the Apple II computer, decided maybe becoming a maker of PCs was a better idea. …
Lenovo Plays The Long Game, Not The Wrong Game, In Systems was written by Timothy Prickett Morgan at The Next Platform.
Arm Is The New RISC/Unix, RISC-V Is The New Arm
When computer architectures change in the datacenter, the attack always comes from the bottom. …
Arm Is The New RISC/Unix, RISC-V Is The New Arm was written by Timothy Prickett Morgan at The Next Platform.
Nvidia’s “Lovelace” GPU Enters The Datacenter Through The Metaverse
Like everyone else on planet Earth, we were expecting for the next generation of graphics cards based on the “Ada Lovelace” architecture to be announced at the GTC fall 2022 conference this week, but we did not expect for the company to deliver a passively cooled, datacenter server friendly variant of the GeForce RTX 6000 series quite so fast. …
Nvidia’s “Lovelace” GPU Enters The Datacenter Through The Metaverse was written by Timothy Prickett Morgan at The Next Platform.
The “Hopper” GPU Compute Ramp Finally Starts
You can’t be certain about a lot of things in the world these days, but one thing you can count on is the voracious appetite for parallel compute, high bandwidth memory, and high bandwidth networking for AI training workloads. …
The “Hopper” GPU Compute Ramp Finally Starts was written by Timothy Prickett Morgan at The Next Platform.
SambaNova Doubles Up Chips To Chase AI Foundation Models
One of the first tenets of machine learning, which is a very precise kind of data analytics and statistical analysis, is that more data beats a better algorithm every time. …
SambaNova Doubles Up Chips To Chase AI Foundation Models was written by Timothy Prickett Morgan at The Next Platform.
Arm Fills In Some Gaps – And Details – In Server Chip Roadmaps
Let’s just say it right here at the beginning. The first wave of attempts at creating Arm server chips –Calxeda, Applied Micro, AMD, Marvell, Nvidia, and Samsung, among others – was disappointing. …
Arm Fills In Some Gaps – And Details – In Server Chip Roadmaps was written by Timothy Prickett Morgan at The Next Platform.
Hyperscalers And Clouds Switch Up To High Bandwidth Ethernet
The hunger for more compute and storage capacity and for more bandwidth to shuffle and shuttle ever-increasing amounts of data is not insatiable among the hyperscalers and large cloud builders of the world. …
Hyperscalers And Clouds Switch Up To High Bandwidth Ethernet was written by Timothy Prickett Morgan at The Next Platform.
Why Aren’t There Software-Defined NUMA Servers Everywhere?
For decades, we have been using software to chop up servers with virtualization hypervisors to run many small workloads on a relatively big piece of iron. …
Why Aren’t There Software-Defined NUMA Servers Everywhere? was written by Timothy Prickett Morgan at The Next Platform.
Broadcom Secures The Edge By Beefing Up The Network Core
A lot of compute is moving to the edge, and that means that networking and storage has to follow it. …
Broadcom Secures The Edge By Beefing Up The Network Core was written by Timothy Prickett Morgan at The Next Platform.
IBM Power10 Shreds Ice Lake Xeons For Transaction Processing
Here is a simple algebraic equation that describes the relative computing oomph of two different CPU architectures over the past two decades: If Intel an X86 core is X, then an IBM Power core equals 2X. …
IBM Power10 Shreds Ice Lake Xeons For Transaction Processing was written by Timothy Prickett Morgan at The Next Platform.
The Bellwethers Of Enterprise IT Spending Fare Reasonably Well
If you want to figure out what is going on with spending on the corporate datacenters of the world, a good place to start is to examine the financial results of Hewlett Packard Enterprise and Dell Technologies, which are still the two largest original equipment makers for servers in the world. …
The Bellwethers Of Enterprise IT Spending Fare Reasonably Well was written by Timothy Prickett Morgan at The Next Platform.