Javier Picorel
YOU?
Author Swipe
View article: Accelerating Transfer Learning with Near-Data Computation on Cloud Object Stores
Accelerating Transfer Learning with Near-Data Computation on Cloud Object Stores Open
Storage disaggregation underlies today's cloud and is naturally complemented by pushing down some computation to storage, thus mitigating the potential network bottleneck between the storage and compute tiers. We show how ML training benef…
View article: SPARTA: A Divide and Conquer Approach to Address Translation for\n Accelerators
SPARTA: A Divide and Conquer Approach to Address Translation for\n Accelerators Open
Virtual memory (VM) is critical to the usability and programmability of\nhardware accelerators. Unfortunately, implementing accelerator VM efficiently\nis challenging because the area and power constraints make it difficult to\nemploy the …
View article: SPARTA: A Divide and Conquer Approach to Address Translation for Accelerators
SPARTA: A Divide and Conquer Approach to Address Translation for Accelerators Open
Virtual memory (VM) is critical to the usability and programmability of hardware accelerators. Unfortunately, implementing accelerator VM efficiently is challenging because the area and power constraints make it difficult to employ the lar…
View article: Enabling Storage Class Memory as a DRAM Replacement for Datacenter Services
Enabling Storage Class Memory as a DRAM Replacement for Datacenter Services Open
With emerging storage-class memory (SCM) nearing commercialization, there is evidence that it will deliver the much-anticipated high density and access latencies within only a few factors of DRAM. Nevertheless, the latency-sensitive nature…
View article: Storage-Class Memory Hierarchies for Scale-Out Servers.
Storage-Class Memory Hierarchies for Scale-Out Servers. Open
With emerging storage-class memory (SCM) nearing commercialization, there is evidence that it will deliver the much-anticipated high density and access latencies within only a few factors of DRAM. Nevertheless, the latency-sensitive nature…
View article: Near-Memory Address Translation
Near-Memory Address Translation Open
Virtual memory (VM) is a crucial abstraction in modern computer systems at any scale, from handheld devices to datacenters. VM provides programmers the illusion of an always sufficiently large and linear memory, making programming easier. …
View article: The Mondrian Data Engine
The Mondrian Data Engine Open
The increasing demand for extracting value out of ever-growing data poses an ongoing challenge to system designers, a task only made trickier by the end of Dennard scaling. As the performance density of traditional CPU-centric architecture…
View article: parsa-epfl/qflex: QFlex Initial Release V1.0
parsa-epfl/qflex: QFlex Initial Release V1.0 Open
Quick & Flexible Computer Architecture Simulation.
View article: Near-Memory Address Translation
Near-Memory Address Translation Open
Memory and logic integration on the same chip is becoming increasingly cost effective, creating the opportunity to offload data-intensive functionality to processing units placed inside memory chips. The introduction of memory-side process…