Applying Learning Techniques to Understand the Write Performance of Large-scale Parallel Filesystems.
Abstract not provided.
Xie, Bing↗
Engineering topics
Carns, Phil
Publications and source records attributed to Carns, Phil.
mochi-quintain
mochi-quintain is a microservice (i.e., a Mochi provider) that responds to parameterized RPCs to generate synthetic concurrent server load.
Carns, Phil↗
mochi-hpc/mochi-margo
Margo provides Argobots-aware bindings to the Mercury RPC library. Margo combines Mercury and Argobots to simplify development of distributed services. Mercury operations are presented as conventional blocking operations, and RPC handlers are presented as sequential threads. This configuration enables high degree of concurrency while hiding the complexity associated with asynchronous communication progress and callback management.
Carns, Phil↗
Py-Bedrock
Py-Bedrock providers Python utilities to configure and deploy Mochi-based services using Python.
Dorier, Matthieu↗
SSG (Scalable Service Groups)
SSG is a group membership microservice based on the Mercury RPC system. It provides mechanisms for bootstrapping sets of Mercury processes into logical groups and for managing the membership of these process groups over time. At a high-level, each group collectively maintains a group view, which is just a mapping from group member identifiers to Mercury address information. The initial group membership view is specified completely when the group is bootstrapped (created).
Carns, Phil↗
ch-placement
libch-placement is a modular consistent hashing library. It can be used by a distributed storage system to access multiple hashing algorithms, distance metrics, and virtual node settings using a consistent API.
Carns, Phil↗
Tau SDSKV Plugin
The TAU SDSKV Plugin logs TAU Profiling events to an SDSKV Key-Value Database.
Carns, Phil↗
REsource Migration Interface
REMI is a Mochi microservice designed to handle the migration of sets of files from a node to another. It uses RDMA and memory mapping to efficiently transfer potentially large groups of files at once.
Dorier, Matthieu↗