Loading...
Search for:
servers
0.01 seconds
Total 115 records
On dynamic assignment of impatient customers to parallel queues
, Article 2003 International Conference on Dependable Systems and Networks, San Francisco, CA, 22 June 2003 through 25 June 2003 ; 2003 , Pages 751-759 ; Sharif University of Technology
2003
Abstract
Consider a number of parallel queues, each having unlimited capacity and multiple identical exponential servers. The service discipline in each queue is first-come-first-served (FCFS). Customers arrive according to a state-dependent Poisson process with an arrival rate which is a function of the total number of customers in the system. Upon arrival, a customer joins a queue according to a state-dependent stationary policy, where the state of the system is taken to be the number of customers in each queue. No jockeying among queues is allowed. Each arriving customer has a generally distributed deadline until the beginning of its service, after which it must depart the system immediately. An...
On queueing with customer impatience until the end of service
, Article Stochastic Models ; Volume 22, Issue 1 , 2006 , Pages 149-173 ; 15326349 (ISSN) ; Sharif University of Technology
2006
Abstract
We study queueing systems where customers have strict deadlines until the end of their service. An analytic method is given for the analysis of a class of such queues, namely, M(n)/M/1 + G models with ordered service. These are single-server queues with state-dependent Poisson arrival process, exponential service times, FCFS service discipline, and general customer impatience. We derive a closed-form solution for the conditional probability density function of the offered sojourn time, given the number of customers in the system. This is a novel result that has not been seen before. Using this result, we show how the probability measure induced by the offered sojourn time is computed, and...
Near-Ideal networks-on-chip for servers
, Article 23rd IEEE Symposium on High Performance Computer Architecture, HPCA 2017, 4 February 2017 through 8 February 2017 ; 2017 , Pages 277-288 ; 15300897 (ISSN); 9781509049851 (ISBN) ; Modarressi, M ; Sarbazi Azad, H ; Sharif University of Technology
IEEE Computer Society
2017
Abstract
Server workloads benefit from execution on many-core processors due to their massive request-level parallelism. A key characteristic of server workloads is the large instruction footprints. While a shared last-level cache (LLC) captures the footprints, it necessitates a low-latency network-on-chip (NOC) to minimize the core stall time on accesses serviced by the LLC. As strict quality-of-service requirements preclude the use of lean cores in server processors, we observe that even state-of-the-art single-cycle multi-hop NOCs are far from ideal because they impose significant NOC-induced delays on the LLC access latency, and diminish performance. Most of the NOC delay is due to per-hop...
An Efficient Approach for Spam Mail Detection in the Sender Side
, M.Sc. Thesis Sharif University of Technology ; Nemaney Pour, Alireza (Supervisor)
Abstract
Spam mails are unwanted mails sent to large numbers of users. Such emails not only consume the network resources but also cause lots of security uncertainties. Based on our observation, the location where the spam filters operate in is an important parameter to save network resources. In this thesis, we introduce a new and effective approach to avoid spam emails from being transferred by relocating the filtering system in the sender mail server. Compared with previous methods in our method, the spam email is detected after spammer clicks on the send button. Here the sender mail server decides based on some pre-defined criteria. If the sender mail server determines that the email is not spam,...
A Novel Reliability Model for Information Technology Web services
, M.Sc. Thesis Sharif University of Technology ; Fatahi Valilai, Omid (Supervisor)
Abstract
The Internet is often used for communication-based applications such as online banking, stock trading, and shopping where service interruption is unacceptable. Thus, analysis of the impact of the failures of hardware, software, and related services’ performance quality of the service provided to users is of importance to designers and managers in these functions. In this study, we developed a strategy to the design of a dynamic combination of file servers with the aim of providing reliability and SLA. At first, we examine the reasons for the violation of service level in file host services and datacenters, and then we introduce the criteria to measure the violation of service level on the...
Power reduction in HPC data centers: a joint server placement and chassis consolidation approach
, Article Journal of Supercomputing ; Vol. 70, issue. 2 , 2014 , p. 845-879 ; Momtazpour, M ; Goudarzi, M ; Sharif University of Technology
Abstract
Size and number of high-performance data centers are rapidly growing all around the world in recent years. The growth in the leakage power consumption of servers along with its exponential dependence on the ever increasing process variation in nanometer technologies has made it inevitable to move toward variation-aware power reduction strategies in data centers. In this paper, we address the problem of joint server placement and chassis consolidation to minimize power consumption of high-performance computing data centers under process variation. To this end, we introduce two variation-aware server placement heuristics as well as an integer linear programming (ILP)-based server placement...
Coded load balancing in cache networks
, Article IEEE Transactions on Parallel and Distributed Systems ; Volume 31, Issue 2 , 2020 , Pages 347-358 ; Parvaresh, F ; Pourmiri, A ; Shariatpanahi, S. P ; Sharif University of Technology
IEEE Computer Society
2020
Abstract
We consider load balancing problem in a cache network consisting of storage-enabled servers forming a distributed content delivery scenario. Previously proposed load balancing solutions cannot perfectly balance out requests among servers, which is a critical issue in practical networks. Therefore, in this paper, we investigate a coded cache content placement where coded chunks of original files are stored in servers based on the files popularity distribution. In our scheme, upon each request arrival at the delivery phase, by dispatching enough coded chunks to the request origin from the nearest servers, the requested file can be decoded. Here, we show that if $n$n requests arrive randomly at...
Non-preemptive earliest-deadline-first scheduling policy: A performance study
, Article MASCOTS 2005: 13th IEEE International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunications Systems, Atlanta, GA, 27 September 2005 through 29 September 2005 ; Volume 2005 , 2005 , Pages 201-208 ; 15267539 (ISSN) ; Movaghar, A ; Sharif University of Technology
2005
Abstract
This paper introduces an analytical method for approximating the performance of a soft real-time system modeled by a single-server queue. The service discipline in the queue is earliest-deadline-first (EDF), which is an optimal scheduling policy. Real-time jobs with exponentially distributed deadlines arrive according to a Poisson process. All jobs have deadlines until the end of service and are served non-preemptively. Occurrences of transient faults in the server are also taken into account. The important performance measure to calculate is the loss probability due to deadline misses and/or transient faults. The system is approximated by a Markovian model in the long run. A key parameter,...
A method for performance analysis of earliest-deadline-first scheduling policy
, Article 2004 International Conference on Dependable Systems and Networks, Florence, 28 June 2004 through 1 July 2004 ; 2004 , Pages 826-834 ; Movaghar, A ; Sharif University of Technology
IEEE Computer Society
2004
Abstract
This paper introduces an analytical method for approximating the fraction of jobs that miss their deadlines in a real-time system when earliest-deadline-first scheduling policy (EDF) is used. In the system, jobs either all have deadlines until the beginning of service or deadlines until the end of service. In the former case, EDF is known to be optimal and, in the latter case, it will be optimal if preemption is allowed. In both cases, the system is modeled by an M/M/1/EDF+M queue, i.e., a single server queue with Poisson arrival, and service times and customer impatience, which are exponentially distributed. The optimality property of EDF is used for the estimation of a key parameter, γn...
A robust and efficient SIP authentication scheme
, Article Scientia Iranica ; Volume 17, Issue 1 D , June , 2010 , Pages 25-38 ; 10263098 (ISSN) ; Darmani, Y ; Jalili, R ; Nourani, M ; Sayad Haghighi, M ; Sharif University of Technology
2010
Abstract
The Session Initiation Protocol (SIP), which is becoming the de facto standard for the next-generation VoIP networks, is currently receiving much attention in many aspects. One aspect that was not deeply addressed in the original SIP is its authentication procedure. Apart from its security, an SIP authentication procedure should be efficient. This paper proposes a robust and efficient three-party SIP authentication protocol. In this protocol, the. end users are. authenticated with the proxy server in their domain using the. registrar server. Compared to previous works, our proposed protocol is more efficient and secure. To support our protocol with a formal security proof, its model is...
Proximity-Aware balanced allocations in cache networks
, Article 31st IEEE International Parallel and Distributed Processing Symposium, IPDPS 2017, 29 May 2017 through 2 June 2017 ; 2017 , Pages 1068-1077 ; 9781538639146 (ISBN) ; Jafari Siavoshan, M ; Shariatpanahi, S. P ; Sharif University of Technology
Institute of Electrical and Electronics Engineers Inc
2017
Abstract
We consider load balancing in a network of caching servers delivering contents to end users. Randomized load balancing via the so-called power of two choices is a well-known approach in parallel and distributed systems that reduces network imbalance. In this paper, we propose a randomized load balancing scheme which simultaneously considers cache size limitation and proximity in the server redirection process. Since the memory limitation and the proximity constraint cause correlation in the server selection process, we may not benefit from the power of two choices in general. However, we prove that in certain regimes, in terms of memory limitation and proximity constraint, our scheme results...
On the security of O-PSI a delegated private set intersection on outsourced datasets
, Article 2017 14th International ISC (Iranian Society of Cryptology) Conference on Information Security and Cryptology, ISCISC 2017 ; 2018 , Pages 77-81 ; 9781538665602 (ISBN) ; Delavar, M ; Ameri, M. H ; Mohajeri, J ; Aref, M. R ; Sharif University of Technology
Abstract
In recent years, determining the common information privately and efficiently between two mutually mistrusting parties have become an important issue in social networks. Many Private set intersection (PSI) protocols have been introduced to address this issue. By applying these protocols, two parties can compute the intersection between their sets without disclosing any information about components that are not in the intersection. Due to the broad range of computational resources that the cloud can provide for its users, determining the set intersection by cloud may decrease the computational cost of the users. The proposed protocol by Abadi et al. is one of the introduced protocols in this...
Storage, communication, and load balancing trade-off in distributed cache networks
, Article IEEE Transactions on Parallel and Distributed Systems ; Volume 29, Issue 4 , April , 2018 , Pages 943-957 ; 10459219 (ISSN) ; Pourmiri, A ; Shariatpanahi, S. P ; Sharif University of Technology
IEEE Computer Society
2018
Abstract
We consider load balancing in a network of caching servers delivering contents to end users. Randomized load balancing via the so-called power of two choices is a well-known approach in parallel and distributed systems. In this framework, we investigate the tension between storage resources, communication cost, and load balancing performance. To this end, we propose a randomized load balancing scheme which simultaneously considers cache size limitation and proximity in the server redirection process. In contrast to the classical power of two choices setup, since the memory limitation and the proximity constraint cause correlation in the server selection process, we may not benefit from the...
Evaluation of hardware data prefetchers on server processors
, Article ACM Computing Surveys ; Volume 52, Issue 3 , 2019 ; 03600300 (ISSN) ; Tabaeiaghdaei, S ; Lotfi Kamran, P ; Sarbazi Azad, H ; Sharif University of Technology
Association for Computing Machinery
2019
Abstract
Data prefetching, i.e., the act of predicting an application's future memory accesses and fetching those that are not in the on-chip caches, is a well-known and widely used approach to hide the long latency of memory accesses. The fruitfulness of data prefetching is evident to both industry and academy: Nowadays, almost every high-performance processor incorporates a few data prefetchers for capturing various access patterns of applications; besides, there is a myriad of proposals for data prefetching in the research literature, where each proposal enhances the efficiency of prefetching in a specific way. In this survey, we evaluate the effectiveness of data prefetching in the context of...
Designing Instruction Prefetcher with Low Area Overhead for Server Workloads
,
M.Sc. Thesis
Sharif University of Technology
;
Sarbazi Azad, Hamid
(Supervisor)
;
Lotfi Kamran, Pejman
(Co-Supervisor)
Abstract
L1 instruction cache misses creates a crucial performance bottleneck for server applications. Server applications extensively use operating system services, and as such, have large instruction footprint that dwarfs instruction cache size. Meanwhile, fast access requirements preclude enlarging instruction cache that can hold the whole instruction footprint of current server workloads. Prior works proposed using hardware prefetching schemes to eliminate or reduce the effect of instruction cache misses. They use the fact that server application instruction sequences are repetitive. So by recording and prefetching based on such sequesnces, L1 insruction misses could be reduced. While they...
An efficient spam mail detection by counter technique
, Article World Academy of Science, Engineering and Technology ; Volume 50 , May , 2011 , Pages 573-576 ; 2010376X (ISSN) ; Behnam Roudsari, S ; Nemaney Pour, A ; Sharif University of Technology
Abstract
Spam mails are unwanted mails sent to large number of users. Spam mails not only consume the network resources, but cause security threats as well. This paper proposes an efficient technique to detect, and to prevent spam mail in the sender side rather than the receiver side. This technique is based on a counter set on the sender server. When a mail is transmitted to the server, the mail server checks the number of the recipients based on its counter policy. The counter policy performed by the mail server is based on some pre-defined criteria. When the number of recipients exceeds the counter policy, the mail server discontinues the rest of the process, and sends a failure mail to sender of...
Flopcoin: a cryptocurrency for computation offloading
, Article IEEE Transactions on Mobile Computing ; 2017 ; 15361233 (ISSN) ; Ahmadi, M ; Kosta, S ; Hui, P ; Sharif University of Technology
Abstract
During the last years, researchers have proposed solutions to help smartphones improve execution time and reduce energy consumption by offloading heavy tasks to remote entities. Lately, inspired by the promising results of message forwarding in opportunistic networks, many researchers have proposed strategies for task offloading towards nearby mobile devices, giving birth to the Device-to-Device offloading paradigm. None of these strategies, though, offers any mechanism that considers selfish users and, most importantly, that motivates and defrays the participating devices who spend their resources. In this paper, we address these problems and propose the design of a framework that...
A mapreduce algorithm for metric anonymity problems
, Article 31st Canadian Conference on Computational Geometry, CCCG 2019, 8 August 2019 through 10 August 2019 ; 2019 , Pages 117-123 ; Ghodsi, M ; Miri, S ; Sharif University of Technology
Canadian Conference on Computational Geometry
2019
Abstract
We focus on two metric clusterings namely r-gather and (r, ?)-gather. The objective of r-gather is to minimize the radius of clustering, such that each cluster has at least r points. (r, ?)-gather is a version of r-gather with the extra condition that at most n? points can be left unclustered (outliers). MapReduce is a model used for processing big data. In each round, it distributes data to multiple servers, then simultaneously processes each server's data. We prove a lower bound 2 on the approximation factor of metric r-gather in the MapReduce model, even if an optimal algorithm for r-gather exists. Then, we give a (4+ δ)-approximation algorithm for r-gather in MapReduce which runs in O(...
A mapreduce algorithm for metric anonymity problems
, Article 31st Canadian Conference on Computational Geometry, CCCG 2019, 8 August 2019 through 10 August 2019 ; 2019 , Pages 117-123 ; Ghodsi, M ; Miri, S ; Sharif University of Technology
Canadian Conference on Computational Geometry
2019
Abstract
We focus on two metric clusterings namely r-gather and (r, ?)-gather. The objective of r-gather is to minimize the radius of clustering, such that each cluster has at least r points. (r, ?)-gather is a version of r-gather with the extra condition that at most n? points can be left unclustered (outliers). MapReduce is a model used for processing big data. In each round, it distributes data to multiple servers, then simultaneously processes each server's data. We prove a lower bound 2 on the approximation factor of metric r-gather in the MapReduce model, even if an optimal algorithm for r-gather exists. Then, we give a (4+ δ)-approximation algorithm for r-gather in MapReduce which runs in O(...
Minimizing expected discounted cost in a queueing loss model with discriminating arrivals
, Article European Journal of Operational Research ; Volume 282, Issue 2 , 2020 , Pages 593-601 ; Ross, S ; Sharif University of Technology
Elsevier B.V
2020
Abstract
We consider a queuing loss system with heterogeneous skill based servers and Poisson arrivals. We first assume that each arrival has a vector (X1,…,Xn) of independent binary random variables with Xi=1 if server i is eligible to serve that arrival. The service time at server i is exponential with rate μi. Arrivals finding no servers that are both idle and eligible to serve them are lost. Assuming the system incurs a cost of one unit for each lost customer, our goal is to find the optimal policy for assigning arrivals to idle and eligible servers so as to minimize the expected discounted cost of the system. Later, we generalize our model by considering k server pools where each pool i is...