Peter R. Pietzuch

Hermes: a distributed event-based middleware architecture

In this paper, we argue that there is a need for an event-based middleware to build large-scale distributed systems. Existing publish/subscribe systems still have limitations compared to invocation-based middlewares. We introduce Hermes, a novel event-based distributed middleware architecture that follows a type- and attribute-based publish/subscribe model. It centres around the notion of an event type and supports features commonly known from object-oriented languages like type hierarchies and super-type subscriptions. A scalable routing algorithm using an overlay routing network is presented that avoids global broadcasts by creating rendezvous nodes. Fault-tolerance mechanisms that can cope with different kinds of failures in the middleware are integrated with the routing algorithm resulting in a scalable and robust system.

Network-Aware Operator Placement for Stream-Processing Systems

To use their pool of resources efficiently, distributed stream-processing systems push query operators to nodes within the network. Currently, these operators, ranging from simple filters to custom business logic, are placed manually at intermediate nodes along the transmission path to meet application-specific performance goals. Determining placement locations is challenging because network and node conditions change over time and because streams may interact with each other, opening venues for reuse and repositioning of operators. This paper describes a stream-based overlay network (SBON), a layer between a stream-processing system and the physical network that manages operator placement for stream-processing systems. Our design is based on a cost space, an abstract representation of the network and on-going streams, which permits decentralized, large-scale multi-query optimization decisions. We present an evaluation of the SBON approach through simulation, experiments on PlanetLab, and an integration with Borealis, an existing stream-processing engine. Our results show that an SBON consistently improves network utilization, provides low stream latency, and enables dynamic optimization at low engineering cost.

Composite event detection as a generic middleware extension

Event-based communication provides a flexible and robust approach to monitoring and managing large-scale distributed systems. Composite event detection extends the scope and flexibility of these systems by allowing application components to express interest in complex patterns of events. This makes it possible to handle the large numbers of events generated in Internet-wide systems, and in network monitoring and pervasive computing applications. In this article, we introduce a novel generic composite event detection framework that can be added on top of existing middleware architectures, as demonstrated in our implementation over JMS. We argue that the framework is flexible, expressive and easy to implement. Based on finite state automata extended with a rich time model and parameterization support, it provides a decomposable core language for specifying composite events. This allows detection to be distributed automatically throughout the system, guided by distribution policies that control the quality of service. Finally, tests show that using our composite event system over JMS can reduce bandwidth consumption while maintaining low notification delay for composite events.

A framework for event composition in distributed systems

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. In this paper, we introduce a general composite event detection framework that can be added on top of existing middleware architectures -- as demonstrated in our implementation over JMS. We argue that the framework is flexible, expressive, and easy to implement. Based on finite state automata extended with a rich time model and support for parameterisation, it provides a decomposable core language for composite event specification, so that composite event detection can be distributed throughout the system. We discuss the issues associated with automatic distribution of composite event expressions. Finally, tests of our composite event system over JMS show reduced bandwidth consumption and a low notification delay for composite events.

Peer-to-peer overlay broker networks in an event-based middleware

Overlay broker networks are an important part of an event-based middleware. In this paper, we investigate the requirements of overlay broker networks and argue that using peer-to-peer techniques for their creation and the content-based routing of events has many advantages. We support our claims with an experimental evaluation of Hermes, an event-based middleware architecture that uses a peer-to-peer routing substrate, in comparison with a standard publish/subscribe system that has a simple, predefined overlay topology. The results reveal that Hermes has better routing efficiency and keeps less routing state at the event brokers.

Distributed complex event processing with query rewriting

The nature of data in enterprises and on the Internet is changing. Data used to be stored in a database first and queried later. Today timely processing of new data, represented as events, is increasingly valuable. In many domains, complex event processing (CEP) systems detect patterns of events for decision making. Examples include processing of environmental sensor data, trades in financial markets and RSS web feeds. Unlike conventional database systems, most current CEP systems pay little attention to query optimisation. They do not rewrite queries to more efficient representations or make decisions about operator distribution, limiting their overall scalability. This paper describes the NEXT CEP system that was especially designed for query rewriting and distribution. Event patterns are specified in a high-level query language and, before being translated into event automata, are rewritten in a more efficient form. Automata are then distributed across a cluster of machines for detection scalability. We present algorithms for query rewriting and distributed placement. Our experiments on the Emulab test-bed show a significant improvement in system scalability due to rewriting and distribution.

Role-based access control for publish/subscribe middleware architectures

Research into publish/subscribe messaging has so far done little to propose architectures for the support of access control, yet this will be an increasingly critical requirement as systems move to Internet-scale. This paper discusses the general requirements of publish/subscribe systems with access control. We then present our specific integration of OASIS role-based access control into the Hermes publish/subscribe middleware platform. Our system supports many advanced features, such as the ability to work within a network where nodes are attributed different levels of trust, and employs a variety of access restriction methods which balance expressiveness with the content-based routing optimisations available. We illustrate our achievements by discussing an application scenario in which our system will be of particular use.

Access control in publish/subscribe systems

Two convincing paradigms have emerged for achieving scalability in widely distributed systems: publish/subscribe communication and role-based, policy-driven control of access to the system by applications. A strength of publish/subscribe is its many-to-many communication paradigm and loose coupling of components, so that publishers need not know the recipients of their data and subscribers need not know the number and location of publishers. But some data is sensitive, and its visibility must be controlled carefully for personal and legal reasons. We describe the requirements of several application domains where the event-based paradigm is appropriate yet where security is an issue. Typical are the large-scale systems required by government and public bodies for domains such as healthcare, police, transport and environmental monitoring. We discuss how a publish/subscribe service can be secured; firstly by specifying and enforcing access control policy at the service API, and secondly by enforcing the security and privacy aspects of these policies within the service network itself. Finally, we describe an alternative to whole-message encryption, appropriate for highly sensitive and long-lived data destined for specific domains with varied requirements. We outline our investigations and findings from several research projects in these areas.

Stable and Accurate Network Coordinates

Network coordinates provide a scalable way to estimate latencies among large numbers of hosts. While there are several algorithms for producing coordinates, none account for the fact that nodes observe a stream of distinct observations that may vary by as much as three orders-ofmagnitude. With such variable data, coordinate systems are prone to high error and instability in live deployments. In addition, dynamics such as triangle violations can lead to coordinate oscillations, producing further instability and making it difficult for applications to know when their coordinates have truly changed. Because simulation results demonstrate that network coordinates are capable of providing low cost and sufficiently accurate answers to common queries, it is vital that we develop the ability to obtain similar results in practice. We propose two filters which combined to improve network coordinate accuracy by 54% and coordinate stability by 96% when run on a real, largescale network.

Information Flow Control for Secure Cloud Computing

Security concerns are widely seen as an obstacle to the adoption of cloud computing solutions. Information Flow Control (IFC) is a well understood Mandatory Access Control methodology. The earliest IFC models targeted security in a centralised environment, but decentralised forms of IFC have been designed and implemented, often within academic research projects. As a result, there is potential for decentralised IFC to achieve better cloud security than is available today. In this paper we describe the properties of cloud computing-Platform-as-a-Service clouds in particular-and review a range of IFC models and implementations to identify opportunities for using IFC within a cloud computing context. Since IFC security is linked to the data that it protects, both tenants and providers of cloud services can agree on security policy, in a manner that does not require them to understand and rely on the particulars of the cloud software stack in order to effect enforcement.