As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
Benchmarking and rating constantly running, distributed, message based systems is very difficult. Time based benchmarks require high effort and are very error prone and expensive because they need dedicated hardware. Instead we propose to use the number of messages in such systems as an abstract layer which is independent from physical clusters but nevertheless allows to infer physical behaviour using hardware specific cost vectors. The number of messages and cost vectors can be systematically extracted from a running instance of a message based system. We show how results gained using a traditional benchmark and those inferred using our technique correlate using two real world examples.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.