Search

This paper considers a stochastic optimization approach for job scheduling and server management in large-scale, geographically distributed data centers. Randomly arriving jobs are routed to a choice of servers. The number of active servers depends on server activation decisions that are updated at a slow time scale, and the service rates of the servers are controlled by power scaling decisions that are made at a faster time scale. We develop a two-time-scale decision strategy that offers provable power cost and delay guarantees. The performance and robustness of the approach is illustrated through simulations. [ABSTRACT FROM AUTHOR]

In the above article , we introduced two scheduling policies and analyzed their average job latencies. With an implicit assumption that the scheduling policies provide sample-path bounds by construction, we claimed that their average job latencies serve as upper and lower bounds on that of a centralized MDS queue. In this note, we present recently discovered counterexamples, disproving the assumption. We replace the assumption with a conjecture that the average latency bounds still hold. We also provide an erratum to the original article to correct any confusing or misleading statements. [ABSTRACT FROM AUTHOR]

In order to scale economically, data centers are increasingly evolving their data storage methods from simple data replication to more powerful erasure codes, which provide the same level of reliability as replication, but at a significantly lower storage cost. In particular, it is well known that maximum-distance-separable (MDS) codes, such as Reed–Solomon codes, can achieve a target reliability with the maximum storage efficiency. While the use of codes for providing improved reliability in archival storage systems, where data is less frequently accessed (or so-called “cold data”), is well understood, the role of codes in storing more frequently accessed and active “hot data”, where latency is the key metric, is less clear. In this paper, we study data storage systems based on MDS codes through the lens of queueing theory, and term the queueing system arising under codes as an “MDS queue.” We provide lower and upper bounds on the average job latency for both centralized and decentralized versions of MDS queues. We also provide extensive simulations to corroborate our analysis as well as obtain additional insights. [ABSTRACT FROM PUBLISHER]

User identification across multiple datasets has a wide range of applications and there has been an increasing set of research works on this topic during recent years. However, most of existing works focus on user identification with a single input data type, e.g., (I) identifying a user across multiple social networks with online data and (II) detecting a single user from heterogeneous trajectory datasets with offline data. Different from previous works, in this paper, we propose a framework on user identification between online and offline datasets. We build connections between these two types of data by a mapping from IP addresses to physical locations. To solve this problem, we propose a novel framework consisting of three steps. First, we use a clustering method based on locations of IP addresses to map IP addresses into specific physical location distributions. Second, we propose a novel pairwise index to reduce space cost and running time for computing the co-occurrence. Lastly, we apply a learning-to-rank method to merge the effect of multiple features we get in the first two steps. Based on our framework, we design experiments to demonstrate the efficiency (in time and space) of our framework, together with the precision and recall of our approach compared to other methods. [ABSTRACT FROM AUTHOR]

This comment points out several mathematical errors in the proof of Therorem 3, and gives the correct expression of B3 . [ABSTRACT FROM AUTHOR]