Time to start writing your paper for the 14th AsHES workshop, to be co-located with @ipdps! Authors are invited to submit research in areas of parallel and distributed processing with focus on emerging heterogeneous platforms! Find the entire CFP in: ashes-hpc.org/2024/cfp.html

Time is ticking! A very interesting venue for impactful work on high performance computing and heterogeneous architectures!

ANNOUNCMENT! Deadline extended to 26th of January! Two weeks to submit your work on AsHES! IPDPS #HeterogeneousSystems #HPC

Remember our work, PARALiA, for multi-GPU BLAS? Here is Petros Anastasiadis from CSLab@NTUA presenting results for communication-optimized, multi-GPU, high-performance, energy-efficient GEMM with PARALiA, at HiPEAC Session 2 on Runtime Optimization!

Deadline approaching in less than a week! Time to finalize and submit your work on AsHES on January 26th!

1/2 Our paper "Elastic Translations: Fast virtual memory with multiple translation sizes" has been accepted at MICRO Symposium! We (Chloe Alverti , CSLab@NTUA) extend Linux to optimally and transparently take advantage of ARMv8-A's OS-assisted TLB coalesecing. #MICRO57 #MICRO2024

Jovan Stojkovic (Jovan Stojkovic ) presenting his research, "Chasing the Tail at Scale: Towards Cloud-Native Processor Architectures", at the Heidelberg Laureate Forum (Heidelberg Laureate Forum). #HLF24 Jovan is a PhD student at @IllinoisCDS advised by Josep Torrellas (jovans2.github.io)

Do not miss this Monday at #MICRO2024 (MICRO Symposium) Stratos talking about his work on Elastic Translations! It will take place at 10:10 at Room A! Come to hear how modern CPUs can break the 2MB/1GiB huge page model!

Great talk yesterday by Stratos Psomadakis on a full room! Come and find him today at the poster session in the morning to learn more about Elastic Translations! #MICRO2024 CSLab@NTUA

Our paper on CXL is now out! We cover: - Latencies/BW for CXL mem-exp - Instruction lat (LD/ST, atomic...) - Microarch utilization - Intel CAT and prefetch - Intel DSA and CXL - Several app: LLMs, vectorDB, graph, KV-stores, and VMs! Check it out on arxiv.org/abs/2411.02814

The program of the 8th Computing Systems Research Day (January 7th, NTUA, Zografou Campus) with Kostis Kaffes Kostis Kaffes Nikos Vasilakis Nikos Vasilakis Chloe Alverti Chloe Alverti Stratos Psomadakis Stratos Psomadakis and Demos Masouros is here: cslab.ece.ntua.gr See you there!

Kostis Kaffes Is it finally time for efficient distributed shared memory? Chloe Alverti Chloe Alverti discussing how CXL can contribute to this ultimate goal!

Kostis Kaffes Chloe Alverti Stratos Psomadakis Stratos Psomadakis presents his work on elastic translations, i.e. how to use multiple page sizes for a single process address space in order to accelerate virtual memory translation

Chloe (Chloe Alverti) and Stratos (Stratos Psomadakis) and Co received a Best Paper Award at ASPLOS for their work -- "CXLfork: Fast Remote Fork over CXL Fabrics." CXLfork is a remote fork interface that realizes close to zero-serialization, zero-copy process cloning across nodes over

#ILLINOIS CS professor Tianjin Xu is seeking to improve the reliability of modern cloud operations to empower the next generation of cloud-computing! Xu shared his visions in the Q&A below ☁️💻 ▶️bit.ly/3XPaOF2