A Competitive Analysis for Balanced Transactional Memory Workloads
نویسندگان
چکیده
منابع مشابه
Workshop on Transactional Memory Workloads
The last half of the 1990's saw a concerted effort to take transactions out of the database and shove them into mass-market programming environments. Both Microsoft and Sun heralded their respective offerings (MTS and EJB) as "the answer." However, by the end of the decade, it was no longer clear that the right question was being asked nor that the average developer liked the answer they were g...
متن کاملTowards Load Balanced Distributed Transactional Memory
We consider the problem of implementing transactional memory in d-dimensional mesh networks. We present and analyze MultiBend, a novel load balanced directory-based protocol, which is designed for the data-flow distributed implementation of software transactional memory. It supports three basic operations, publish, lookup, and move, on a shared object. A pleasing aspect of MultiBend is that it ...
متن کاملTransactional Scheduling for Read-Dominated Workloads
The transactional approach to contention management guarantees atomicity by aborting transactions that may violate consistency. A major challenge in this approach is to schedule transactions in a manner that reduces the total time to perform all transactions (the makespan), since transactions are often aborted and restarted. The performance of a transactional scheduler can be evaluated by the r...
متن کاملLightweight Transactional Arrays for Read-Dominated Workloads
Many common workloads rely on arrays as a basic data structure on top of which they build more complex behavior. Others use them because they are a natural representation for their problem domains. Software Transactional Memory (STM) has been proposed as a new concurrency control mechanism that simplifies concurrent programming. Yet, most STM implementations have no special representation for a...
متن کاملExploring Parallelism in Transactional Workloads
Multicores are now standard in most machines, which means that many programmers are faced with the challenge of how to take advantage of all the potential parallelism. Transactional Memory (TM) promises to simplify this task. Yet, at the same time, TM inhibits the programmer from fully exploring the latent parallelism in his application. In particular, it does not allow a transaction to contain...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Algorithmica
سال: 2011
ISSN: 0178-4617,1432-0541
DOI: 10.1007/s00453-011-9532-3