iManage: Policy-Driven Self-Management of Enterprise-Scale Systems
Online Demo
It is obvious that big, complex enterprise systems are hard to manage. What is not obvious is how to make them more manageable. Although there is a growing body of research into system self-management, many techniques are either too narrow, focusing on a single component rather than the entire system, or not robust enough, failing to scale or respond to the full range of an administrator's needs. In our iManage system we have developed a policy-driven system modeling framework that aims to bridge the gap between manageable components and manageable systems. In particular, iManage provides: (1) system state-space partitioning, which divides a large system state-space into partitions that are more amenable to constructing system models and developing policies, (2) online model and policy adaptation to allow the self-management infrastructure to deal gracefully with changes in operating environment, system configuration, and workload, and (3) tractability and trust, where tractability allows an administrator to understand why the system chose a particular policy and also influence that decision, and trust allows an administrator to understand the system's confidence in a proposed, automated action.
iFLOW: Distributed Data-Stream Management using Utility-Driven Self-Adaptive Middleware
Online Demo
We consider enterprise-wide information flows that are responsible for acquiring, processing and delivering operational information across the business units. Middleware that enables such aggregation of data-streams must not only support scalable and efficient self-management to deal with changes in the operating conditions, but should also have an embedded sense of value of the data to appreciate the critical nature of some updates. iFLOW is an autonomic middleware for implementing enterprise-wide information in a utility-aware self-managing way. iFLOW offers a clean abstraction of an information flow and is therefore capable of implementing several other messaging models including the Enterprise information flows. iFLOW, in essence, is a utility-aware self-managing implementation of a general information flow abstraction. The iManage project listed above resulted from the self-management lessons that we learnt from this work.
Data Dissemination Tools & Techniques
Distributed Stream Query Processing
The project focused on using a SQL-like language for expressing the stream composition requirements. As part of this project we focused on optimizing the deployment of stream queries such that the total cost imposed by such a query per unit of time was reduced. We also looked at the problem of optimization in presence of multiple queries with a focus on reuse of existing query operators. This work later evolved into the much larger iFLOW project which is a middleware for support general information flows.
IQ-GridFTP & High-Performance Data-Overlays
The IQ-GridFTP project modeified the popular GridFTP software to provide a modified FTP protocol that enabled client-requested filtering or modification of file content at the FTP server before it was transmitted to the client. This proves to be very useful in grid like environment where file sizes may reach many gigabytes. The High-Performance Data-Overlays work focused on utilizing multiple internet paths between a server and a client to achieve maximum throughput. As opposed to other available techniques the throttling in our approach is based both on the characteristics of the network paths and the data-content. We later developed algorithms that calculated overlay path guarantees like "Path A can consistently provide X Mbps with a guarantee of P percent".
The project focused on using a SQL-like language for expressing the stream composition requirements. As part of this project we focused on optimizing the deployment of stream queries such that the total cost imposed by such a query per unit of time was reduced. We also looked at the problem of optimization in presence of multiple queries with a focus on reuse of existing query operators. This work later evolved into the much larger iFLOW project which is a middleware for support general information flows.
IQ-GridFTP & High-Performance Data-Overlays
The IQ-GridFTP project modeified the popular GridFTP software to provide a modified FTP protocol that enabled client-requested filtering or modification of file content at the FTP server before it was transmitted to the client. This proves to be very useful in grid like environment where file sizes may reach many gigabytes. The High-Performance Data-Overlays work focused on utilizing multiple internet paths between a server and a client to achieve maximum throughput. As opposed to other available techniques the throttling in our approach is based both on the characteristics of the network paths and the data-content. We later developed algorithms that calculated overlay path guarantees like "Path A can consistently provide X Mbps with a guarantee of P percent".