In a Big data computing, the processing of data requires a large amount of CPU cycles and
network bandwidth and disk I/O. Dataflow is a programming model for processing Big data …

Cloud providers have recently introduced new offerings whereby spare computing
resources are accessible at discounts compared to on-demand computing. Exploiting such …

Worldwide sensor streams are expanding continuously with unbounded velocity in volume,
and for this acceleration, there is an adaptation of large stream data processing system from …

We introduce Harmony, a new scheduling framework that executes multiple Parameter-
Server ML training jobs together to improve cluster resource utilization. Harmony …

With the increase in amount of Big Data being generated each year, tools and technologies
developed and used for the purpose of storing, processing and analyzing Big Data has also …

(57) ABSTRACT A lifecycle management method, system, and computer program product
include establishing a public key infra structure (PKI) for end-to-end encryption of control …

2019-03-21 Assigned to INTERNATIONAL BUSINESS MACHINES CORPORATION
reassignment INTERNATIONAL BUSINESS MACHINES CORPORATION ASSIGNMENT OF …

Model parallelism is a standard paradigm to decouple a deep neural network (DNN) into
sub-nets when the model is large. Recent advances in class parallelism significantly reduce …

Robustness is a fundamental and timeless issue, and it remains vital to all aspects of
computation systems, regardless of specific computation platforms, architectures, and …

We build a working system that implements our approaches. The above two solutions are
implemented in the same system and share the runtime part that can dynamically migrate …