Optimizing data access latencies in cloud systems by intelligent virtual machine placement

Abstract

Many cloud applications are data intensive requiring the processing of large data sets and the MapReduce/Hadoop architecture has become the de facto processing framework for these applications. Large data sets are stored in data nodes in the cloud which are typically SAN or NAS devices. Cloud applications process these data sets using a large number of application virtual machines (VMs), with the total completion time being an important performance metric. There are many factors that affect the total completion time of the processing task such as the load on the individual servers, the task scheduling mechanism, communication and data access bottlenecks, etc. One dominating factor that affects completion times for data intensive applications is the access latencies from processing nodes to data nodes. Ideally, one would like to keep all data access local to minimize access latency but this is often not possible due to the size of the data sets, capacity constraints in processing nodes which constrain VMs from being placed in their ideal location and so on. When it is not possible to keep all data access local, one would like to optimize the placement of VMs so that the impact of data access latencies on completion times is minimized. We address this problem of optimized VM placement -given the location of the data sets, we need to determine the locations for placing the VMs so as to minimize data access latencies while satisfying system constraints. We present optimal algorithms for determining the VM locations satisfying various constraints and with objectives that capture natural tradeoffs between minimizing latencies and incurring bandwidth costs. We also consider the problem of incorporating inter-VM latency constraints. In this case, the associated location problem is NPhard with no effective approximation within a factor of 2 − for any > 0. We discuss an effective heuristic for this case and evaluate by simulation the impact of the various tradeoffs in the optimization objectives.

Rigorous results on the effectiveness of some heuristics for the consolidation of virtual machines in a cloud data center

“…On the other hand, attacking the whole problem at once may be too difficult because the search space is so huge; in this respect, a decomposition approach may be preferable. Similar questions arise also in the context of other decompositions of the VM allocation problem: e.g., whether data nodes should be placed first and compute nodes only afterwards [26] or the placement of data nodes and compute nodes should be optimized together [31]. This is a highly nontrivial trade-off that must be resolved for every optimization problem and every potential decomposition approach separately.…”

“…Breitgand and Epstein presented a 2-approximation algorithm for the stochastic bin packing problem under the assumption of independent normally distributed random variables [24]. Alicherry and Lakshman derived some approximation algorithms and inapproximability results for the problem of minimizing the cost of communication among VMs [25,26]. Breitgand et al devised algorithms for profit optimization in a federated cloud and proved that, under certain conditions, the algorithm for one of the sub-problems, which is a greedy LProunding procedure, ensures 2-approximation [27].…”

Security Policy Scheme for an Efficient Security Architecture in Software-Defined Networking

“…Furthermore, MOVR manages and optimizes the resource use of cloud applications. Aicherry et al proposed optimal algorithms to solve the problem of optimized virtual machine (VM) placement when the location of the data sets is given [9]. The proposed algorithms try to capture the best tradeoff point between minimizing latencies and incurring bandwidth costs.…”

Allocation of Virtual Machines in Cloud Data Centers—A Survey of Problem Models and Optimization Algorithms

“…For some restricted problem versions, polynomial-time approximation algorithms have been presented with rigorously proven approximation guarantees [1,2,18,68].…”