Peter Pietzuch | Imperial College London (original) (raw)

Papers by Peter Pietzuch

Abstract This paper describes XenoTrust, the trust management architecture used in the XenoServer... more Abstract This paper describes XenoTrust, the trust management architecture used in the XenoServer Open Platform: a public infrastructure for wide-area computing, capable of hosting tasks that span the full spectrum of distributed paradigms. We suggest that using an event-based publish/subscribe methodology for the storage, retrieval and aggregation of reputation information can help exploiting asynchrony and simplicity, as well as improving scalability.

Stream-based overlay networks (SBONs) are one approach to implementing large-scale stream process... more Stream-based overlay networks (SBONs) are one approach to implementing large-scale stream processing systems. A fundamental consideration in an SBON is that of service placement, which determines the physical location of in-network processing services or operators, in such a way that network resources are used efficiently.

Abstract Complex middleware frameworks are made out of interacting components which may include b... more Abstract Complex middleware frameworks are made out of interacting components which may include bugs. These frameworks are often extended to provide additional features by third-party extensions that may not be completely trusted and, as a result, compromise the security of the whole platform. Aiming to minimize these problems, we propose a demonstration of PrivateFlow, a publish/subscribe prototype supported by Decentralized Information Flow Control (DIFC).

Abstract Cloud computing is gaining acceptance among mainstream technology users. Storage cloud p... more Abstract Cloud computing is gaining acceptance among mainstream technology users. Storage cloud providers often employ Storage Area Networks (SANs) to provide elasticity, rapid adaptability to changing demands, and policy based automation. As storage capacity grows, the storage environment becomes heterogeneous, increasingly complex, harder to manage, and more expensive to operate.

Abstract For large-scale distributed applications such as internet-wide or ubiquitous systems, ev... more Abstract For large-scale distributed applications such as internet-wide or ubiquitous systems, event-based communication is an effective messaging mechanism between components. In order to handle the large volume of events in such systems, composite event detection enables application components to express interest in the occurrence of complex patterns of events.

Abstract—Social networking applications on the web handle the personal data of a large number of ... more Abstract—Social networking applications on the web handle the personal data of a large number of concurrently active users. These applications must comply with complex privacy requirements, while achieving scalability and high performance. Applying constraints to the flow of data through such applications to enforce privacy policy is challenging because individual components process data belonging to many different users.

Abstract—A modular design is proposed and analyzed for accelerating the publish/subscribe matchin... more Abstract—A modular design is proposed and analyzed for accelerating the publish/subscribe matching algorithm in reconfigurable hardware. With help from a performance model, we demonstrate an optimized FPGA implementation which is scalable and efficient enough for many of today's most demanding web and financial applications. Our design achieves 5.9 times speedup over software while consuming around 0.5% of power.

Abstract The rapid proliferation of cloud and service-oriented computing infrastructure is creati... more Abstract The rapid proliferation of cloud and service-oriented computing infrastructure is creating an ever increasing thirst for storage within data centers. Ideally management applications in cloud deployments should operate in terms of high-level goals, and not present specific implementation details to administrators. Cloud providers often employ Storage Area Networks (SANs) to gain storage scalability.

Abstract Today we lack an infrastructure for globally processing stream data from sensor networks... more Abstract Today we lack an infrastructure for globally processing stream data from sensor networks and making this data available to millions of users in real-time. To build such a system, we need to address a set of challenges and, in particular, rethink what dependability means in this context: it is infeasible to guarantee perfect data processing at a global scale. Instead, the degradation of result quality due to failure and resource shortages should be made explicit to users.

Abstract The emergence of sensor networks and distributed applications that generate data streams... more Abstract The emergence of sensor networks and distributed applications that generate data streams has created a need for Internet overlays designed for streaming data. Such stream-based overlay network (SBONs) consist of a set of Internet hosts that collect, process, and deliver stream-based data to multiple applications. A key challenge in the design and implementation of SBONs is efficient path optimization when mapping logical query streams to physical network hosts and paths.

Abstract Event-based systems give the potential for active information sharing. The event-based p... more Abstract Event-based systems give the potential for active information sharing. The event-based paradigm, if used for event transport, provides loose coupling between components, many-to-many communication and mutual anonymity of event producers and event consumers. This communication style has been adopted enthusiastically for convenience of programming; particularly for financial processing, healthcare applications and sensor-based systems.

Abstract In this paper we present an analysis of using run-time reconfiguration of reconfigurable... more Abstract In this paper we present an analysis of using run-time reconfiguration of reconfigurable hardware to modify trading algorithms during use. This provides flexibility in algorithm design, enabling the implementation to be reactive to changes in market conditions, increasing in performance. We study what can be achieved to reduce performance loss in algorithms while reconfiguration takes place, such as buffering information during this time.

Abstract The publish/subscribe (pub/sub) communications paradigm is suitable for building large-s... more Abstract The publish/subscribe (pub/sub) communications paradigm is suitable for building large-scale, widely distributed applications. Distributed pub/sub middleware scales well because it decouples communicating clients. However, complete decoupling of clients make it more challenging to design distributed applications using pub/sub middleware: often clients want some information about each other.

Security engineering must be integrated with all stages of application specification and developm... more Security engineering must be integrated with all stages of application specification and development to be effective. Doing this properly is increasingly critical as organisations rush to offload their software services to cloud providers. Service-level agreements (SLAs) with these providers currently focus on performance-oriented parameters, which runs the risk of exacerbating an impedance mismatch with the security middleware.

Abstract Routing substrates for overlay networks are an important building block for large distri... more Abstract Routing substrates for overlay networks are an important building block for large distributed applications. Many existing substrates are based on a random identifier space and therefore do not respect node locality when routing data. This can lead to lower performance for locality-sensitive applications, such as web caching, distributed gaming, and resource discovery.

Abstract Intelligent workload consolidation in storage systems leads to better Return On Investme... more Abstract Intelligent workload consolidation in storage systems leads to better Return On Investment (ROI), in terms of more efficient use of data center resources, better Quality of Service (QoS), and lower power consumption. This is particularly significant yet challenging in a cloud environment, in which a large set of different workloads multiplex on a shared, heterogeneous infrastructure. However, the increasing availability of fine grained workload logging facilities allows better insights to be gained from workload profiles.

Internet applications that route data over default Internet paths can often increase performance ... more Internet applications that route data over default Internet paths can often increase performance by sending their traffic over alternative “detour” paths. Previous work has shown that applications can use detour routing to improve end-to-end metrics such as latency and path availability. However, the potential of detour routing has yet to be applied where it may be most important: improving TCP throughput. In this paper, we study the feasibility of bandwidth detouring on the Internet.

Abstract Large-scale distributed applications need latency information to make network-aware rout... more Abstract Large-scale distributed applications need latency information to make network-aware routing decisions. Collecting these measurements, however, can impose a high burden. Network coordinates are a scalable and efficient way to supply nodes with up-to-date latency estimates. We present our experience of maintaining network coordinates on PlanetLab.

ABSTRACT Stream processing systems are becoming increasingly important to analyse real-time data ... more ABSTRACT Stream processing systems are becoming increasingly important to analyse real-time data generated by modern applications such as online social networks. Their main characteristic is to produce a continuous stream of fresh results as new data are being generated at real-time. Resource provisioning of stream processing systems is difficult due to time-varying workload data that induce unknown resource demands over time.

Abstract This paper describes XenoTrust, the trust management architecture used in the XenoServer... more Abstract This paper describes XenoTrust, the trust management architecture used in the XenoServer Open Platform: a public infrastructure for wide-area computing, capable of hosting tasks that span the full spectrum of distributed paradigms. We suggest that using an event-based publish/subscribe methodology for the storage, retrieval and aggregation of reputation information can help exploiting asynchrony and simplicity, as well as improving scalability.

Stream-based overlay networks (SBONs) are one approach to implementing large-scale stream process... more Stream-based overlay networks (SBONs) are one approach to implementing large-scale stream processing systems. A fundamental consideration in an SBON is that of service placement, which determines the physical location of in-network processing services or operators, in such a way that network resources are used efficiently.

Abstract Complex middleware frameworks are made out of interacting components which may include b... more Abstract Complex middleware frameworks are made out of interacting components which may include bugs. These frameworks are often extended to provide additional features by third-party extensions that may not be completely trusted and, as a result, compromise the security of the whole platform. Aiming to minimize these problems, we propose a demonstration of PrivateFlow, a publish/subscribe prototype supported by Decentralized Information Flow Control (DIFC).

Abstract Cloud computing is gaining acceptance among mainstream technology users. Storage cloud p... more Abstract Cloud computing is gaining acceptance among mainstream technology users. Storage cloud providers often employ Storage Area Networks (SANs) to provide elasticity, rapid adaptability to changing demands, and policy based automation. As storage capacity grows, the storage environment becomes heterogeneous, increasingly complex, harder to manage, and more expensive to operate.

Abstract For large-scale distributed applications such as internet-wide or ubiquitous systems, ev... more Abstract For large-scale distributed applications such as internet-wide or ubiquitous systems, event-based communication is an effective messaging mechanism between components. In order to handle the large volume of events in such systems, composite event detection enables application components to express interest in the occurrence of complex patterns of events.

Abstract—Social networking applications on the web handle the personal data of a large number of ... more Abstract—Social networking applications on the web handle the personal data of a large number of concurrently active users. These applications must comply with complex privacy requirements, while achieving scalability and high performance. Applying constraints to the flow of data through such applications to enforce privacy policy is challenging because individual components process data belonging to many different users.

Abstract—A modular design is proposed and analyzed for accelerating the publish/subscribe matchin... more Abstract—A modular design is proposed and analyzed for accelerating the publish/subscribe matching algorithm in reconfigurable hardware. With help from a performance model, we demonstrate an optimized FPGA implementation which is scalable and efficient enough for many of today's most demanding web and financial applications. Our design achieves 5.9 times speedup over software while consuming around 0.5% of power.

Abstract The rapid proliferation of cloud and service-oriented computing infrastructure is creati... more Abstract The rapid proliferation of cloud and service-oriented computing infrastructure is creating an ever increasing thirst for storage within data centers. Ideally management applications in cloud deployments should operate in terms of high-level goals, and not present specific implementation details to administrators. Cloud providers often employ Storage Area Networks (SANs) to gain storage scalability.

Abstract Today we lack an infrastructure for globally processing stream data from sensor networks... more Abstract Today we lack an infrastructure for globally processing stream data from sensor networks and making this data available to millions of users in real-time. To build such a system, we need to address a set of challenges and, in particular, rethink what dependability means in this context: it is infeasible to guarantee perfect data processing at a global scale. Instead, the degradation of result quality due to failure and resource shortages should be made explicit to users.

Abstract The emergence of sensor networks and distributed applications that generate data streams... more Abstract The emergence of sensor networks and distributed applications that generate data streams has created a need for Internet overlays designed for streaming data. Such stream-based overlay network (SBONs) consist of a set of Internet hosts that collect, process, and deliver stream-based data to multiple applications. A key challenge in the design and implementation of SBONs is efficient path optimization when mapping logical query streams to physical network hosts and paths.

Abstract Event-based systems give the potential for active information sharing. The event-based p... more Abstract Event-based systems give the potential for active information sharing. The event-based paradigm, if used for event transport, provides loose coupling between components, many-to-many communication and mutual anonymity of event producers and event consumers. This communication style has been adopted enthusiastically for convenience of programming; particularly for financial processing, healthcare applications and sensor-based systems.

Abstract In this paper we present an analysis of using run-time reconfiguration of reconfigurable... more Abstract In this paper we present an analysis of using run-time reconfiguration of reconfigurable hardware to modify trading algorithms during use. This provides flexibility in algorithm design, enabling the implementation to be reactive to changes in market conditions, increasing in performance. We study what can be achieved to reduce performance loss in algorithms while reconfiguration takes place, such as buffering information during this time.

Abstract The publish/subscribe (pub/sub) communications paradigm is suitable for building large-s... more Abstract The publish/subscribe (pub/sub) communications paradigm is suitable for building large-scale, widely distributed applications. Distributed pub/sub middleware scales well because it decouples communicating clients. However, complete decoupling of clients make it more challenging to design distributed applications using pub/sub middleware: often clients want some information about each other.

Security engineering must be integrated with all stages of application specification and developm... more Security engineering must be integrated with all stages of application specification and development to be effective. Doing this properly is increasingly critical as organisations rush to offload their software services to cloud providers. Service-level agreements (SLAs) with these providers currently focus on performance-oriented parameters, which runs the risk of exacerbating an impedance mismatch with the security middleware.

Abstract Routing substrates for overlay networks are an important building block for large distri... more Abstract Routing substrates for overlay networks are an important building block for large distributed applications. Many existing substrates are based on a random identifier space and therefore do not respect node locality when routing data. This can lead to lower performance for locality-sensitive applications, such as web caching, distributed gaming, and resource discovery.

Abstract Intelligent workload consolidation in storage systems leads to better Return On Investme... more Abstract Intelligent workload consolidation in storage systems leads to better Return On Investment (ROI), in terms of more efficient use of data center resources, better Quality of Service (QoS), and lower power consumption. This is particularly significant yet challenging in a cloud environment, in which a large set of different workloads multiplex on a shared, heterogeneous infrastructure. However, the increasing availability of fine grained workload logging facilities allows better insights to be gained from workload profiles.

Internet applications that route data over default Internet paths can often increase performance ... more Internet applications that route data over default Internet paths can often increase performance by sending their traffic over alternative “detour” paths. Previous work has shown that applications can use detour routing to improve end-to-end metrics such as latency and path availability. However, the potential of detour routing has yet to be applied where it may be most important: improving TCP throughput. In this paper, we study the feasibility of bandwidth detouring on the Internet.

Abstract Large-scale distributed applications need latency information to make network-aware rout... more Abstract Large-scale distributed applications need latency information to make network-aware routing decisions. Collecting these measurements, however, can impose a high burden. Network coordinates are a scalable and efficient way to supply nodes with up-to-date latency estimates. We present our experience of maintaining network coordinates on PlanetLab.

ABSTRACT Stream processing systems are becoming increasingly important to analyse real-time data ... more ABSTRACT Stream processing systems are becoming increasingly important to analyse real-time data generated by modern applications such as online social networks. Their main characteristic is to produce a continuous stream of fresh results as new data are being generated at real-time. Resource provisioning of stream processing systems is difficult due to time-varying workload data that induce unknown resource demands over time.