Tpp: Transparent page placement for cxl-enabled tiered-memory
The increasing demand for memory in hyperscale applications has led to memory becoming
a large portion of the overall datacenter spend. The emergence of coherent interfaces like …
a large portion of the overall datacenter spend. The emergence of coherent interfaces like …
Memory disaggregation: advances and open challenges
H Al Maruf, M Chowdhury - ACM SIGOPS Operating Systems Review, 2023 - dl.acm.org
Compute and memory are tightly coupled within each server in traditional datacenters.
Large-scale datacenter operators have identified this coupling as a root cause behind …
Large-scale datacenter operators have identified this coupling as a root cause behind …
A survey of software techniques to emulate heterogeneous memory systems in high-performance computing
Heterogeneous memory will be involved in several upcoming platforms on the way to
exascale. Combining technologies such as HBM, DRAM and/or NVDIMM allows to tackle …
exascale. Combining technologies such as HBM, DRAM and/or NVDIMM allows to tackle …
Shiftsreduce: Minimizing shifts in racetrack memory 4.0
Racetrack memories (RMs) have significantly evolved since their conception in 2008,
making them a serious contender in the field of emerging memory technologies. Despite key …
making them a serious contender in the field of emerging memory technologies. Despite key …
Performance evaluation of intel optane memory for managed workloads
S Akram - ACM Transactions on Architecture and Code …, 2021 - dl.acm.org
Intel Optane memory offers non-volatility, byte addressability, and high capacity. It suits
managed workloads that prefer large main memory heaps. We investigate Optane as the …
managed workloads that prefer large main memory heaps. We investigate Optane as the …
[HTML][HTML] H2M: exploiting heterogeneous shared memory architectures
J Klinkenberg, A Kozhokanova, C Terboven… - Future Generation …, 2023 - Elsevier
Over the past decades, the performance gap between the memory subsystem and compute
capabilities continued to spread. However, scientific applications and simulations show …
capabilities continued to spread. However, scientific applications and simulations show …
Profiling dynamic data access patterns with controlled overhead and quality
Modern workloads tend to have huge working sets and low locality. Despite this trend, the
capacity of DRAM has not been increased enough to accommodate such huge working sets …
capacity of DRAM has not been increased enough to accommodate such huge working sets …
MaPHeA: A lightweight memory hierarchy-aware profile-guided heap allocation framework
Hardware performance monitoring units (PMUs) are a standard feature in modern
microprocessors for high-performance computing (HPC) and embedded systems, by …
microprocessors for high-performance computing (HPC) and embedded systems, by …
Daos: Data access-aware operating system
In data-intensive workloads, data placement and memory management are inherently
difficult: the programmer and the operating system have to choose between (combinations …
difficult: the programmer and the operating system have to choose between (combinations …
ecoHMEM: Improving object placement methodology for hybrid memory systems in HPC
Recent byte-addressable persistent memory (PMEM) technology offers capacities
comparable to storage devices and access times much closer to DRAMs than other non …
comparable to storage devices and access times much closer to DRAMs than other non …