Austin Donnelly

Austin Donnelly is a Research Software Development Engineer (RSDE) in the Systems group at Microsoft Research, Cambridge. He obtained his B.A. in Computer Science from the University of Cambridge in 1996, and went on to complete his Ph.D. there in 2002.

His most recent interests are in enterprise storage, from both a power-management and performance perspective. He helped build a system which can redirect disk writes across a network to allow disks to be spun down for extended periods of time, saving power.

He has worked on the Seaweed project, building a query infrastructure designed for very large datasets distributed over thousands to millions of machines. An example such dataset might result from the Anemone project, where he implemented an endsystem-based network management system.

Previously, he supported work using Magpie for performance analysis of the distributed system underlying MSN Search.

Before this he worked on Ethernet topology discovery, using the hosts in the network to send probe packets and record where they arrive, to infer where switches, hubs, wireless access points and wireless bridges might be.

His thesis dissertation (titled Resource Control in Network Elements [PS], [PDF]) described how operating systems for active network nodes could provide quality of service guarantees to sandboxed code in an efficient manner, and demonstrated this by way of a prototype implementation.

While a Ph.D. student, Austin worked on the Pegasus II project as part of a team porting the Nemesis research OS to Intel-based PCs. He helped design and implemented the network stack for Nemesis.

Austin spent three months working for AT&T's Florham Park research labs in New Jersey, implementing a standards-compliant video streaming and caching system.

You can contact him by emailing austind (at) microsoft (dot) com, or by telephone on +44 1223 479769.

• Write Off-Loading: Practical Power Management for Enterprise Storage
  [ 1.3MB PDF ]

  Dushyanth Narayanan, Austin Donnelly, Antony Rowstron

  Status:To appear in 6th USENIX Conference on File and Storage Technologies (FAST'08), February 2008

  
  
• Delay-aware querying with Seaweed
  [ Springer abstract page ]

  Dushyanth Narayanan, Austin Donnelly, Richard Mortier, Antony Rowstron

  Status: Published in The VLDB Journal Special Issue: Best of VLDB 2006, September 2007

  
  
• Delay-aware querying with Seaweed
  [ 455KB PDF ]

  Dushyanth Narayanan, Austin Donnelly, Richard Mortier, Antony Rowstron

  Status: Published in 32nd Intl. Conference on Very Large Databases (VLDB 2006), September 2006

  
  
• Reclaiming Network-wide Visibility Using Ubiquitous End System Monitors
  [ 78KB PDF ]

  Evan Cooke, Richard Mortier, Austin Donnelly, Paul Barham, Rebecca Isaacs

  Network-centric tools like NetFlow and security systems like IDSes provide essential data about the availability, reliability, and security of network devices and applications. However, the increased use of encryption and tunnelling has reduced the visibility of monitoring applications into packet headers and payloads (e.g. 93% of traffic on our enterprise network is IPSec encapsulated). The result is the inability to collect the required information using network-only measurements. To regain the lost visibility we propose that measurement systems must themselves apply the end-to-end principle: only endsystems can correctly attach semantics to traffic they send and receive. We present such an end-to-end monitoring platform that ubiquitously records per-flow data and then we show that this approach is feasible and practical using data from our enterprise network.

  Status: Published in USENIX 2006 Annual Technical Conference, June 2006.

  
  
• Seaweed: Distributed Scalable Ad Hoc Querying
  [ 44KB PDF]

  Richard Mortier, Dushyanth Narayanan, Austin Donnelly, Antony Rowstron

  Many emerging applications such as wide-area network management need to query large, structured, highly distributed datasets. Seaweed is a distributed scalable infrastructure for querying such datasets. In this paper we describe its architecture and design features, using the Anemone network management system as a motivating example. The main contribution is a design supporting accurate query planning and efficient execution across a large number of unreliable endsystems. In contrast to prior work, Seaweed supports ad hoc querying in addition to continuous querying. The paper describes the solutions adopted by Seaweed: latency-based cost estimation, availability-based scheduling, and meta-data aggregation.

  Status: Published in Proceedings of 2nd IEEE International Workshop on Networking Meets Databases (NetDB'06), April 2006

  
  
• Fast Scalable Robust Node Enumeration
  [ Springer abstract page ]

  Richard Black, Austin Donnelly, Alexandru Gavrilescu, Dave Thaler

  In a Local Area Network of computers, often a machine wants to learn of the existence of all the others satisfying some condition. Specifically, there are a number of existing discovery algorithms which permit an enumerator to reliably discover protocol participants, many of them idealised. This paper provides a new technique which controls the load placed on the network, minimises the time to completion, handles networks with significant loss, and scales over many orders of magnitude. Most significantly, the protocol also deals with the possibility of a malicious enumerator; an important contribution needed for current real-world networks. We also address the effects of several systems and engineering aspects, including scheduler jitter and clock quantisation

  Status: Published in IFIP Networking 2005, May 2005.

  
  
• Using Magpie for Request Extraction and Workload Modelling
  [ 934KB PDF ]

  Paul Barham, Austin Donnelly, Rebecca Isaacs, Richard Mortier

  Tools to understand complex system behaviour are essential for many performance analysis and debugging tasks, yet there are many open research problems in their development. Magpie is a toolchain for automatically extracting a system's workload under realistic operating conditions. Using low-overhead instrumentation, we monitor the system to record fine-grained events generated by kernel, middleware and application components. The Magpie request extraction tool uses an application-specific event schema to correlate these events, and hence precisely capture the control flow and resource consumption of each and every request. By removing scheduling artefacts, whilst preserving causal dependencies, we obtain canonical request descriptions from which we can construct concise workload models suitable for performance prediction and change detection. In this paper we describe and evaluate the capability of Magpie to accurately extract requests and construct representative models of system behaviour.

  Status: Published in OSDI 2004, December 2004.

  
  
• Ethernet Topology Discovery without Network Assistance
  [ 323KB PDF ]

  Richard Black, Austin Donnelly, Cédric Fournet

  This work addresses the problem of Layer 2 topology discovery. Current techniques concentrate on using SNMP to query information from Ethernet switches. In contrast, we present a technique that infers the Ethernet (Layer 2) topology without assistance from the network elements by injecting suitable probe packets from the end-systems and observing where they are delivered. We describe the algorithm, formally characterize its correctness and completeness, and present our implementation and experimental results. Performance results show that although originally aimed at the home and small office the techniques scale to much larger networks.

  Status: Published in ICNP 2004, October 2004.

  
  
• Lightweight Thread Tunnelling in Network Applications
  [ 126KB gzipped PostScript ], [ 164KB PDF ]

  Austin Donnelly

  Active Network nodes are increasingly being used for non-trivial processing of data streams. These complex network applications typically benefit from protection between their components for fault-tolerance or security. However, fine-grained memory protection introduces bottlenecks in communication among components. This paper describes memory protection in Expert, an OS for programmable network elements which re-examines thread tunnelling as a way of allowing these complex applications to be split over multiple protection domains. We argue that previous problems with tunnelling are symptoms of overly general designs, and we demonstrate a minimal domain-crossing primitive which nevertheless achieves the majority of benefits possible from tunnelling.

  Status: Published in IWAN2002, December 2002.

  
  
• Resource Control in Network Elements
  [ 389KB gzipped PostScript ], [ 965KB PDF ]

  Austin Donnelly

  Status: Dissertation approved for the degree of Doctor of Philosophy at the University of Cambridge, January 2002.

  Also available as Cambridge University Computer Laboratory Technical Report UCAM-CL-TR-534 in PDF.

  
  
• IP Route Lookups as String Matching
  [ 41KB gzipped PostScript ], [ 109KB PDF ]

  Austin Donnelly and Tim Deegan

  An IP route lookup can be considered as a string matching problem on the destination address. Finite State Automata (FSA) are a flexible and efficient way to match strings. This paper describes how a routing table can be encoded as an FSA and how, through a process of state reduction, we can obtain an optimal representation. This gives insights into the basic properties of the longest-prefix match problem.

  Status: Published in IEEE LCN'2000, November 2000.

• A Network-Based Replay Portal.
  [ 93KB gzipped PostScript ], [ 338KB PDF ]

  J. E. van der Merwe, C. J. Sreenan, A. N. Donnelly, A. Basso, C. R. Kalmanek.

  Technologies based on cable modems currently use the capacity of a single TV channel to offer 25 - 30 Mb/s downstream for Internet access. With the advent of Digital TV and the significant bandwidth savings it gleans from video compression, it is expected that providers will increase the access capacity available for IP traffic, while retaining the bulk of the bandwidth for the primary service of broadcast TV. Motivated by the increasing popularity of on-demand streaming media, our work questions this IP-versus-broadcast distinction, and proposes a hybrid model which combines the familiar broadcast model with the conveniences of on-demand TV viewing over IP. In this paper we discuss the potential benefits of this approach, and some of the difficult technical challenges it raises. We propose an architecture of network-based portals and sketch the types of services we envisage it will enable. Further, we describe the design and implementation of a replay service that operates within this architecture to offer a new on-demand TV-viewing experience.

  Status: Published in Proceedings of 6th IFIP Conference on Intelligence in Networks (SmartNet 2000), September 2000.

  
  
• Protocol Implementation in a Vertically Structured Operating System
  [ 52KB gzipped PostScript ], [ 101KB PDF ]

  Richard Black, Paul Barham, Austin Donnelly, Neil Stratford

  A vertically structured Operating System is one in which neither the "kernel" nor "servers" perform work on behalf of applications - the former because it exists only to multiplex the CPU, and the latter in order to avoid Quality of Service interference between the applications. Instead, wherever possible, the applications perform all of their own processing. Such a vertical structure provides many advantages for applications but leads to some interesting problems and opportunities for protocol stack implementation. This paper describes the techniques we used in our protocol implementation and the benefits that the vertical structure provided.

  Status: Published in LCN'97