Kenneth Salem

Adaptive control of virtualized resources in utility computing environments

Data centers are often under-utilized due to over-provisioning as well as time-varying resource demands of typical enterprise applications. One approach to increase resource utilization is to consolidate applications in a shared infrastructure using virtualization. Meeting application-level quality of service (QoS) goals becomes a challenge in a consolidated environment as application resource needs differ. Furthermore, for multi-tier applications, the amount of resources needed to achieve their QoS goals might be different at each tier and may also depend on availability of resources in other tiers. In this paper, we develop an adaptive resource control system that dynamically adjusts the resource shares to individual tiers in order to meet application-level QoS goals while achieving high resource utilization in the data center. Our control system is developed using classical control theory, and we used a black-box system modeling approach to overcome the absence of first principle models for complex enterprise applications and systems. To evaluate our controllers, we built a testbed simulating a virtual data center using Xen virtual machines. We experimented with two multi-tier applications in this virtual data center: a two-tier implementation of RUBiS, an online auction site, and a two-tier Java implementation of TPC-W. Our results indicate that the proposed control system is able to maintain high resource utilization and meets QoS goals in spite of varying resource demands from the applications.

Main memory database systems: an overview

Main memory database systems (MMDBs) store their data in main physical memory and provide very high-speed access. Conventional database systems are optimized for the particular characteristics of disk storage mechanisms. Memory resident systems, on the other hand, use different optimizations to structure and organize data, as well as to make it reliable. The authors survey the major memory residence optimizations and briefly discuss some of the MMDBs that have been designed or implemented. >

Adaptive block rearrangement

An adaptive technique for reducing disk seek times is described. The technique copies frequently referenced blocks from their original locations to reserved space near the middle of the disk. Reference frequencies need not be known in advance. Instead, they are estimated by monitoring the stream of arriving requests. Trace-driven simulations show that seek times can be cut substantially by copying only a small number of blocks using this technique. The technique has been implemented by modifying a UNIX device driver. No modifications are required to the file system that uses the driver.

How to roll a join: asynchronous incremental view maintenance

Incremental refresh of a materialized join view is often less expensive than a full, non-incremental refresh. However, it is still a potentially costly atomic operation. This paper presents an algorithm that performs incremental view maintenance as a series of small, asynchronous steps. The size of each step can be controlled to limit contention between the refresh process and concurrent operations that access the materialized view or the underlying relations. The algorithm supports point-in-time refresh, which allows a materialized view to be refreshed to any time between the last refresh and the present.

Altruistic locking

Long-lived transactions (LLTs) hold on to database resources for relatively long periods of time, significantly delaying the completion of shorter and more common transactions. To alleviate this problem we propose an extension to two-phase locking, called altruistic locking, whereby LLTs can release their locks early. Transactions that access this released data are said to run in the wake of the LLT and must follow special locking rules. Like two-phase locking, altruistic locking is easy to implement and guarantees serializability.

Checkpointing memory-resident databases

A database system is considered in which a main-memory database system holds all data in semiconductor memory, and for recovery purposes a backup copy of the database is maintained in secondary storage. The checkpointer is the component of the crash recovery manager responsible for maintaining the backup copy. Ideally, the checkpointer should maintain an almost-up-to-date backup while interfering as little as possible with the systems transaction processing activities. Several algorithms for maintaining such a backup database are presented and compared using an analytic model. The results show some significant performance differences among the algorithms and illustrate some of the tradeoffs that are available in designing such a checkpointer.<<ETX>>

Coordinating activities through extended sagas: a summary

The authors describe an environment designed to support activities such as a purchase order. They propose a simple set of services which would be useful for describing and executing activities. In an implementation, an underlying system would provide these services for activities, much as an operating system provides a set of services for processes. The environment consists of a system call interface (Create, Bind, Commit, Abort, CompensationBind, Send, Receive) which programs can use to request services. These services are designed to support some of the important requirements of data processing activities, including concurrency, modularity, fault tolerance, rollback, and communication.<<ETX>>

Lazy database replication with ordering guarantees

Lazy replication is a popular technique for improving the performance and availability of database systems. Although there are concurrency control techniques, which guarantee serializability in lazy replication systems, these techniques result in undesirable transaction orderings. Since transactions may see stale data, they may be serialized in an order different from the one in which they were submitted. Strong serializability avoids such problems, but it is very costly to implement. We propose a generalized form of strong serializability that is suitable for use with lazy replication. In addition to having many of the advantages of strong serializability, it can be implemented more efficiently. We show how generalized strong serializability can be implemented in a lazy replication system, and we present the results of a simulation study that quantifies the strengths and limitations of the approach.

System M: a transaction processing testbed for memory resident data

System M is an experimental transaction processing testbed that runs on top of the Mach operating system. Its database is stored in primary memory. The structure and algorithms used in System M are described. The checkpointer is the component that periodically sweeps memory and propagates updates to a backup database copy on disk. Several different checkpointing (and logging) algorithms were implemented, and their performance was experimentally evaluated. >

Automatic virtual machine configuration for database workloads

Virtual machine monitors are becoming popular tools for the deployment of database management systems and other enterprise software. In this article, we consider a common resource consolidation scenario in which several database management system instances, each running in a separate virtual machine, are sharing a common pool of physical computing resources. We address the problem of optimizing the performance of these database management systems by controlling the configurations of the virtual machines in which they run. These virtual machine configurations determine how the shared physical resources will be allocated to the different database system instances. We introduce a virtualization design advisor that uses information about the anticipated workloads of each of the database systems to recommend workload-specific configurations offline. Furthermore, runtime information collected after the deployment of the recommended configurations can be used to refine the recommendation and to handle changes in the workload. To estimate the effect of a particular resource allocation on workload performance, we use the query optimizer in a new what-if mode. We have implemented our approach using both PostgreSQL and DB2, and we have experimentally evaluated its effectiveness using DSS and OLTP workloads.