Abstract: A method comprises receiving, by a workload profiler, a representative workload of a computing system under analysis. The workload profiler determines a workload profile of the computing system that reflects a transaction mix that varies over times. A capacity analyzer receives the workload profile, and determines a maximum capacity of the computing system under analysis for serving the workload profile while satisfying a defined quality of service (QoS) target.

Abstract: In one embodiment of the present invention, a method for measuring the capacity of a media server for supporting concurrent media streams to a population of clients is provided. The method comprises identifying a mixed workload, wherein the mixed workload corresponds to a number of media streams served concurrently by the media server to a population of clients wherein all of said number of media streams do not comprise common content encoded at a common bit rate. The method further comprises using a cost function to compute a load for the media server under the mixed workload.

Abstract: An HTTP server is programmed to detect timed-out client requests by inferring states of its client-server connections. These connections, being full-duplex, can each be viewed as including a client-to-server channel and a server-to-client channel. The state of the server-to-client channel can be inferred by examining local server information to determine whether the client-to-server channel is still established. The server processes a request by a client if the inferred state indicates that the server-to-client channel is still established, and the server terminates the client request if the inferred state indicates that the server-to-client channel is no longer established. Consequently, the server does not expend resources by processing timed-out or dead client requests.

Abstract: Systems and methods for efficiently distributing a file from a first node to a plurality of recipient nodes are provided. A method comprises distributing a plurality of subfiles that comprise a file from a first node to a first group comprising a plurality of recipient nodes, wherein at least one subfile distributed from the first node to each recipient node of the first group but no individual recipient node receives all of the plurality of subfiles from the first node. The method further comprises exchanging subfiles among the plurality of recipient nodes of the first group, wherein at least one recipient node of the first group begins communicating a portion of the subfile that it is receiving from the first node to at least one other recipient node before the at least one recipient node fully receives the subfile.

Abstract: A method comprises determining an amount to which resource usage by a driver domain on behalf of a given virtual machine (VM) is to be limited. The method further comprises controlling the resource usage by the driver domain on behalf of the given VM so as not to exceed the determined amount. A system comprises a VM, a driver domain, and a shared resource that is usable by the VM and by the driver domain. The system further comprises a controller for limiting usage of the shared resource by the driver domain on behalf of the VM to prevent the usage of the shared resource by the driver domain on behalf of the VM from exceeding a determined amount.

Abstract: A method comprises receiving into a planning tool a representative workload for a consumer. The method further comprises receiving into the planning tool quality of service desires of the consumer which define permitted degraded performance. In certain embodiments, the permitted degraded performance is time-limited wherein demands of the representative workload may exceed a pre-defined utilization constraint for at least one resource servicing the demands for no more than a pre-defined amount of contiguous time. The planning tool determines an allocation of demand of the consumer for each of a plurality of different classes of service (COSs). In certain embodiments, a first COS provides guaranteed resource access for servicing demand allocated thereto, and a second COS provides non-guaranteed resource access for servicing demand allocated thereto. In certain embodiments, the allocation of demand to the different COSs may be determined for both a normal mode and a failure mode of operation.

Abstract: A method comprises observing communication through a virtualized interface between at least one virtual machine (VM) and a driver domain. Based on the observed communication through the virtualized interface, a weighted amount of communication between the VM and the driver domain is determined for each of the VM(s), wherein the weighted amount is based at least in part on at least one characteristic of the observed communication, such as directional flow and/or size of packets in the communication. Another method comprises evaluating CPU utilization of a target network driver domain in processing communications having different characteristics, and deriving corresponding weighting for the different communication characteristics. The derived weighting for the different communication characteristics are employed within a communication monitor that observes communication between virtual machines and the target network driver domain.

Abstract: A method comprises determining an amount of central processing unit (CPU) usage by a driver domain on behalf of a given virtual machine (VM) over a measurement period. The method further comprises determining a share of CPU to schedule for the given VM for a plurality of future scheduling intervals based at least in part on the determined amount of CPU usage by the driver domain on behalf of the given VM over the measurement period. The method further comprises scheduling the determined share of CPU access to the given VM for the plurality of future scheduling intervals.

Abstract: A method comprises determining an aggregate usage of CPU attributable to a given virtual machine (VM), wherein the aggregate usage of CPU attributable to the given VM includes a determined amount of CPU usage by a driver domain, such as by a network driver domain, that is attributable to the given VM. The method further comprises scheduling, by an aggregate proportional-share scheduler, CPU access to the given VM based at least in part on the determined aggregate usage of CPU attributable to the given VM.

Abstract: A method comprises observing communication through a virtualized interface between at least one virtual machine (VM) and a driver domain. The method further comprises determining for each of the at least one VM, based on the observed communication through the virtualized interface, an amount of communication between the VM and the driver domain. In certain embodiments, the method further comprises determining for each of the at least one VM, based on the determined amount of communication between the VM and the driver domain, CPU utilization of the driver domain attributable to the VM. In certain embodiments, the driver domain comprises a network driver domain. In certain embodiments, the driver domain comprises an isolated network diver domain.

Abstract: A method comprises determining a flexible limit on an amount of resource usage by a driver domain on behalf of a given virtual machine (VM). The method further comprises controlling the resource usage by the driver domain on behalf of the given VM so as not to exceed the flexible limit except under a permitted condition. In certain embodiments the resource usage by the driver domain on behalf of the given VM is controlled so as not to exceed the flexible limit except when a slack share of resource usage is available to the driver domain. Such a slack share of resource usage is a share of resource usage allocated to the driver domain that will not otherwise be used on behalf of another VM.

Abstract: A method comprises scheduling, by a weighted proportional-share virtual time processor scheduler, processor access to a first consumer during a time period in which a second consumer is blocked from processor access. When the second consumer becomes unblocked from processor access, the weighted proportional-share virtual time processor scheduler determines whether the length of the time period over which the second consumer was blocked from processor access exceeds a predefined time threshold. When determined that the length of the time period over which the second consumer was blocked from processor access does not exceed the predefined time threshold, then scheduler does not update the virtual time of the second consumer before scheduling processor access for the second consumer.

Abstract: A system comprises a workload manager evaluator operable to receive a representative workload that is representative of competing workloads that share access to at least one shared computing resource. The workload manager evaluator is operable to evaluate performance of a scheduler that schedules access of the competing workloads to the shared computing resource according to defined control parameter values, wherein the workload manager evaluator evaluates performance of the scheduler under the representative workload for a plurality of different values of the control parameters. In certain embodiments, the workload manager evaluator determines an optimal value for the control parameters of the scheduler for scheduling access to the at least one shared computing resource for the representative workload to satisfy defined performance desires of the system.

Abstract: A system comprises a workload evaluator that is operable to receive a representative workload that is representative of competing demands for capacity of at least one shared computing resource. The workload evaluator evaluates the representative workload and computes a metric representing a degree of burstiness of demands present in the representative workload. The metric representing degree of burstiness of the representative workload may be used for estimating an upper bound on quality of service provided by a workload manager to the representative workload. The metric may also be used for evaluating at least one scheduler parameter setting of the workload manager to aid in determining an optimal parameter setting based at least in part on the estimated impact of the representative workload on QoS provided by the workload manager.

Abstract: A method comprises defining a scheduler parameter for a maximum allocation of capacity of a shared resource to a consumer for a scheduling interval. Utilization of an allocated capacity of the shared resource by the consumer during a given scheduling interval is measured, and when the allocated capacity of the shared resource is completely utilized by the consumer during the given scheduling interval, the scheduler increases the allocated capacity of the shared resource to the defined maximum allocation for the consumer for a next scheduling interval. Thus, rather than gradually increasing the allocation of capacity over many intervals, the scheduler immediately increases the allocation to a predefined maximum amount in response to an allocated amount of capacity being completely utilized during a scheduling interval.

Abstract: In certain embodiments, a method for modeling the memory of a streaming media server is provided. The method comprises creating a segment-based access model for at least one streaming media file available on a streaming media server, and using the segment-based access model to construct a segment-based model of the media server's memory. In certain embodiments, a system comprises a server comprising a memory having a capacity, the server operable to serve at least one streaming file to clients communicatively coupled thereto. The system further comprises a controller operable to determine an amount of unique bytes of the at least one streaming file served by the server during an elapsed time interval, wherein the amount of unique bytes are equal to the capacity of the server's memory.

Abstract: According to one embodiment of the present invention, a method for reconstructing client web page accesses is provided that comprises capturing network-level information for client accesses of at least one web page. The method further comprises constructing a knowledge base of reliable information identifying content of at least one web page, and reconstructing at least one client web page access from the captured network-level information using the reliable information of the knowledge base.

Abstract: A method of distributing a file from a first node to a plurality of recipient nodes comprises logically organizing a plurality of recipient nodes to which a file is to be distributed into a plurality of groups of recipient nodes. The file is distributed to the plurality of groups, wherein the distributing includes (a) partitioning the file into a plurality of subfiles, (b) distributing the plurality of subfiles from a first node to a first group, wherein at least one subfile is distributed from the first node to each recipient node of the first group but not all of the plurality of subfiles are distributed from the first node to any of the recipient nodes of the first group, and (c) the plurality of recipient nodes of the first group exchanging their respective subfiles such that each recipient node of the first group obtains all of the plurality of subfiles.

Abstract: A system and method for distributing a file from a first node to a plurality of recipient nodes are provided. The method comprises partitioning a file into a plurality of subfiles, and distributing the plurality of subfiles from a first node to a first group comprising a plurality of recipient nodes, wherein at least one subfile is distributed from the first node to each recipient node of the first group but not all of the plurality of subfiles are distributed from the first node to any of the recipient nodes of the first group. The method further comprises the plurality of recipient nodes of the first group exchanging their respective subfiles such that each recipient node of the first group obtains all of the plurality of subfiles.

Abstract: According to one embodiment, a method comprises receiving into a planning tool a representative workload for a consumer. The method further comprises determining, by the planning tool, an allocation of demand of the consumer for each of a plurality of different classes of service (COSs). According to one embodiment, a method comprises defining a plurality of classes of service (COSs) for use by a scheduler in allocating capacity of a resource pool to a consumer, wherein the COSs each specify a different priority for accessing the capacity of the resource pool. The method further comprises evaluating, by a planning tool, a representative workload of the consumer, and determining, by the planning tool, a partitioning of resource demands of the representative workload between the plurality of COSs.