Advanced

Reducing Memory in Software-Based Thread-Level Speculation for JavaScript Virtual Machine Execution of Web Applications

Martinsen, Jan Kasper; Grahn, Håkan; Isberg, Anders and Sundström, Henrik (2014) HPCC-2014 the 16th IEEE International Conf. on High Performance and Communications In [Host publication title missing]
Abstract
Thread-Level Speculation has been used to take advantage

of multicore processors in virtual execution environments

for the sequential JavaScript scripting language. While the results

are promising the memory overhead is high. Here we propose to

reduce the memory usage by limiting the checkpoint depth based

on an in-depth study of the memory and execution time effects.

We also propose an adaptive heuristic to dynamically adjust the

checkpoints. We evaluate this using 15 web applications on an

8-core computer. The results show that the memory overhead is

reduced for Thread-Level Speculation by over 90% as compared

to storing all checkpoints. Further,... (More)
Thread-Level Speculation has been used to take advantage

of multicore processors in virtual execution environments

for the sequential JavaScript scripting language. While the results

are promising the memory overhead is high. Here we propose to

reduce the memory usage by limiting the checkpoint depth based

on an in-depth study of the memory and execution time effects.

We also propose an adaptive heuristic to dynamically adjust the

checkpoints. We evaluate this using 15 web applications on an

8-core computer. The results show that the memory overhead is

reduced for Thread-Level Speculation by over 90% as compared

to storing all checkpoints. Further, the performance is often better

than when storing all the checkpoints and at worst 4% slower. (Less)
Please use this url to cite or link to this publication:
author
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
in
[Host publication title missing]
publisher
IEEE--Institute of Electrical and Electronics Engineers Inc.
conference name
HPCC-2014 the 16th IEEE International Conf. on High Performance and Communications
external identifiers
  • Scopus:84929088227
project
EASE
language
English
LU publication?
no
id
77a771e6-16ab-4802-a984-6bb3b159703b (old id 5104391)
alternative location
http://www.jankaspermartinsen.com/dreng/hpcc_jkm_v6.pdf
date added to LUP
2015-02-24 13:38:29
date last changed
2016-06-29 09:02:59
@misc{77a771e6-16ab-4802-a984-6bb3b159703b,
  abstract     = {Thread-Level Speculation has been used to take advantage<br/><br>
of multicore processors in virtual execution environments<br/><br>
for the sequential JavaScript scripting language. While the results<br/><br>
are promising the memory overhead is high. Here we propose to<br/><br>
reduce the memory usage by limiting the checkpoint depth based<br/><br>
on an in-depth study of the memory and execution time effects.<br/><br>
We also propose an adaptive heuristic to dynamically adjust the<br/><br>
checkpoints. We evaluate this using 15 web applications on an<br/><br>
8-core computer. The results show that the memory overhead is<br/><br>
reduced for Thread-Level Speculation by over 90% as compared<br/><br>
to storing all checkpoints. Further, the performance is often better<br/><br>
than when storing all the checkpoints and at worst 4% slower.},
  author       = {Martinsen, Jan Kasper and Grahn, Håkan and Isberg, Anders and Sundström, Henrik},
  language     = {eng},
  publisher    = {ARRAY(0x8561718)},
  series       = {[Host publication title missing]},
  title        = {Reducing Memory in Software-Based Thread-Level Speculation for JavaScript Virtual Machine Execution of Web Applications},
  year         = {2014},
}