IISWC 2015

Final Submission Deadline
April 17, 2015

Submitted by Jaewoong Sim

2015 IEEE International Symposium on Workload Characterization (IISWC)
Atlanta, GA, USA
October (Dates: TBD), 2015
– Abstracts Submission: April 10, 2015
– Paper Submission: April 17, 2015
– Acceptance Notification: June 20, 2015

This symposium is dedicated to the understanding and characterization of
workloads that run on all types of computing systems. New applications and
programming paradigms continue to emerge rapidly as the diversity and
performance of computers increase. On one hand, improvements in computing
technology are usually based on a solid understanding and analysis of existing
workloads. On the other hand, computing workloads evolve and change with
advances in microarchitecture, compilers, programming languages, and networking
communication technologies. Whether they are smart phones and deeply embedded
systems at the low end or massively parallel systems at the high end, the
design of future computing machines can be significantly improved if we
understand the characteristics of the workloads that are expected to run on
them. This symposium will focus on characterizing and understanding emerging
applications in consumer, commercial and scientific computing.

We solicit papers in all areas related to characterization of computing system
workloads. Topics of interest include (but are not limited to):

– Characterization of applications in areas including
o Search engines, e-commerce, web services, databases, application servers
o Embedded, mobile, multimedia, real-time, 3D-Graphics, gaming, telepresence
o Life sciences, bioinformatics, scientific computing, finance, forecasting
o Machine Learning, Analytics, Data mining
o Security, reliability, biometrics
o Grid and Cloud computing
o Emerging big data applications

– Characterization of OS, Virtual Machine, middleware and library behavior
o Virtual machines, Websphere, .NET, Java VM, databases
o Graphics libraries, scientific libraries

– Characterization of system behavior, including
o Operating system and hypervisor effects and overheads
o Hardware accelerators (GPGPU, XML, crypto, etc)
o User behavior and system-user interaction
o Impacts of scale-up and scale-out of systems, applications, and inputs
o Instrumentation methodologies for workload verification and characterization
o Techniques for accurate analysis/measurement of production systems

– Implications of workloads in design including issues, such as
o Power management, reliability, security, performance
o Processors, memory hierarchy, I/O, and networks
o Design of accelerators, FPGA’s, GPU’s, etc.
o Novel architectures (non-Von-Neumann)

– Benchmark creation, analysis, and evaluation issues, including
o Multithreaded benchmarks, benchmark cloning
o Profiling, trace collection, synthetic traces
o Validation of benchmarks

– Analytical and abstract modeling of program behavior and systems

– Emerging and future workloads
o Transactional memory workloads; workloads for multi/many-core system
o Stream-based computing workloads; web2.0/internet workloads;
cyber-physical workloads

– Workloads and characterizations for emerging architectures
o Near data processing architectures
o Heterogeneous memory hierarchies
o New memory hierarchies using die stacked DRAM
o Memory centric architectures with computing in storage, e.g., SSD and disk

General Chair:
– Tom Conte, Georgia Institute of Technology

Program Co-Chairs:
– Hyesoon Kim, Georgia Institute of Technology
– Sudhakar Yalamanchili, Georgia Institute of Technology