The programmable data plane: Abstractions, architectures, algorithms, and applications
Programmable data plane technologies enable the systematic reconfiguration of the low-
level processing steps applied to network packets and are key drivers toward realizing the …
level processing steps applied to network packets and are key drivers toward realizing the …
Direct access,{High-Performance} memory disaggregation with {DirectCXL}
New cache coherent interconnects such as CXL have recently attracted great attention
thanks to their excellent hardware heterogeneity management and resource disaggregation …
thanks to their excellent hardware heterogeneity management and resource disaggregation …
A unified architecture for accelerating distributed {DNN} training in heterogeneous {GPU/CPU} clusters
Data center clusters that run DNN training jobs are inherently heterogeneous. They have
GPUs and CPUs for computation and network bandwidth for distributed training. However …
GPUs and CPUs for computation and network bandwidth for distributed training. However …
{AIFM}:{High-Performance},{Application-Integrated} far memory
Memory is the most contended and least elastic resource in datacenter servers today.
Applications can use only local memory—which may be scarce—even though memory …
Applications can use only local memory—which may be scarce—even though memory …
Netcache: Balancing key-value stores with fast in-network caching
We present NetCache, a new key-value store architecture that leverages the power and
flexibility of new-generation programmable switches to handle queries on hot items and …
flexibility of new-generation programmable switches to handle queries on hot items and …
Clio: A hardware-software co-designed disaggregated memory system
Memory disaggregation has attracted great attention recently because of its benefits in
efficient memory utilization and ease of management. So far, memory disaggregation …
efficient memory utilization and ease of management. So far, memory disaggregation …
Shenango: Achieving high {CPU} efficiency for latency-sensitive datacenter workloads
Datacenter applications demand microsecond-scale tail latencies and high request rates
from operating systems, and most applications handle loads that have high variance over …
from operating systems, and most applications handle loads that have high variance over …
Efficient memory disaggregation with infiniswap
Memory-intensive applications suffer large performance loss when their working sets do not
fully fit in memory. Yet, they cannot leverage otherwise unused remote memory when paging …
fully fit in memory. Yet, they cannot leverage otherwise unused remote memory when paging …
Datacenter {RPCs} can be general and fast
It is commonly believed that datacenter networking software must sacrifice generality to
attain high performance. The popularity of specialized distributed systems designed …
attain high performance. The popularity of specialized distributed systems designed …
Offloading distributed applications onto smartnics using ipipe
Emerging Multicore SoC SmartNICs, enclosing rich computing resources (eg, a multicore
processor, onboard DRAM, accelerators, programmable DMA engines), hold the potential to …
processor, onboard DRAM, accelerators, programmable DMA engines), hold the potential to …