Research Outputs
Permanent URI for this communityhttps://hdl.handle.net/20.500.14288/2
Browse
6 results
Search Results
Publication Restricted An algorithmic framework for instantaneous convolutive bounded component analysis(Koç University, 2014) İnan, Hüseyin Atahan; Erdoğan, Alper Tunga; 0000-0003-0876-2897; Koç University Graduate School of Sciences and Engineering; Electrical and Electronics Engineering; 41624Publication Restricted Energy cost optimization in large scale distributed systems by resource allocation technique(Koç University, 2013) Güler, Hüseyin; Özkasap, Öznur; 0000-0003-4343-0986; Koç University Graduate School of Sciences and Engineering; Computer Science and Engineering; 113507Publication Restricted KUDA: Accelerating dynamic race detection using parallelism on a GPU(Koç University, 2013) Bekar, Ümit Can; Taşıran, Serdar; Koç University Graduate School of Sciences and Engineering; Computer Science and EngineeringPublication Restricted Multi-hop cluster and LTE based heterogeneous architecture for VANET(Koç University, 2013) Uçar, Seyhan; Özkasap, Öznur; 0000-0003-4343-0986; Koç University Graduate School of Sciences and Engineering; Computer Science and Engineering; 113507
Publication Metadata only On the optimal control of parallel processing networks with resource collaboration and multitasking(INFORMS Institute for Operations Research and the Management Sciences, 2021) N/A; Özkan, Erhun; Faculty Member; School of Medicine; 289255We study scheduling control of parallel processing networks in which some resources need to simultaneously collaborate to perform some activities and some resources multitask. Resource collaboration and multitasking give rise to synchronization constraints in resource scheduling when the resources are not divisible, that is, when the resources cannot be split. The synchronization constraints affect the system performance significant-ly. For example, because of those constraints, the system capacity can be strictly less than the capacity of the bottleneck resource. Furthermore, the resource scheduling decisions are not trivial under those constraints. For example, not all static prioritization policies retain the maximum system capacity, and the ones that retain the maximum system capacity do not necessarily minimize the delay (or, in general, the holding cost). We study optimal scheduling control of a class of parallel networks and propose a dynamic prioritization policy that retains the maximum system capacity and is asymptotically optimal in diffusion scale and a conventional heavy-traffic regime with respect to the expected discounted total holding cost objective.Publication Metadata only Trade: precise dynamic race detection for scalable transactional memory systems(Association for Computing Machinery, 2015) Kestor, Gökçen; Ünsal, Osman S.; Cristal, Adrian; Department of Computer Engineering; Taşıran, Serdar; Faculty Member; Department of Computer Engineering; College of Engineering; N/AAs other multithreaded programs, transactional memory (TM) programs are prone to race conditions. Previous work focuses on extending existing definitions of data race for lock-based applications to TM applications, which requires all transactions to be totally ordered "as if" serialized by a global lock. This approach poses implementation constraints on the STM that severely limits TM applications' performance. This article shows that forcing total ordering among all running transactions, while sufficient, is not necessary. We introduce an alternative data race definition, relaxed transactional data race, that requires ordering of only conflicting transactions. The advantages of our relaxed definition are twofold: First, unlike the previous definition, this definition can be applied to a wide range of TMs, including those that do not enforce transaction total ordering. Second, within a single execution, it exposes a higher number of data races, which considerably reduces debugging time. Based on this definition, we propose a novel and precise race detection tool for C/C++ TM applications (TRADE), which detects data races by tracking happens-before edges among conflicting transactions. Our experiments reveal that TRADE precisely detects data races for STAMP applications running on modern STMs with overhead comparable to state-of-the-art race detectors for lock-based applications. Our experiments also show that in a single run, TRADE identifies several races not discovered by 10 separate runs of a race detection tool based on the previous data race definition.