Papers by Debmalya Biswas
Data & Knowledge Engineering, 2009
We consider hierarchical systems where nodes represent entities and edges represent binary relati... more We consider hierarchical systems where nodes represent entities and edges represent binary relationships among them. An example is a hierarchical composition of Web services where the nodes denote services and edges represent the parent-child relationship of a service invoking another service. A fundamental issue to address in such systems is, for two nodes X and Y in the hierarchy whether X can see Y , that is, whether X has visibility over Y . In a general setting, X seeing Y may depend on (i) X wishing to see Y , (ii) Y wishing to be seen by X, and (iii) other nodes not objecting to X seeing Y . The visibility could be with respect to certain attributes like operational details, execution logs, security related issues, etc. In this paper, we develop a generic conceptual model to express visibility. We study two complementary notions: sphere of visibility of a node X that includes all the nodes in the hierarchy that X can see; and sphere of noticeability of X that includes all the nodes that can see X. We also identify dual properties, coherence and correlation, that relate the visibility and noticeability notions, and study their variants.
Peer to Peer (P2P) communities (or “interest groups”) are referred to as nodes that share a commo... more Peer to Peer (P2P) communities (or “interest groups”) are referred to as nodes that share a common interest. Each peer in the system claims to have some interests and, accordingly, would like to become a member of these groups. The available interest groups are arranged according to a hierarchical semantics ontology, and managed with a semantic overlay network. P2P community structure is highly dynamic: a peer may be added to or deleted from a community; communities may be added or deleted; communities may be merged or split; and sub-communities may become parent-level communities and vice versa. In this paper, we propose a highly flexible multi-level data structure to capture the visibility aspect of P2P communities. The data structure is simple, facilitates dynamic changes easily and efficiently in a decentralized fashion, and is highly scalable.
Over the years, the notion of transactions has become synonymous with providing fault-tolerance, ... more Over the years, the notion of transactions has become synonymous with providing fault-tolerance, reliability and robustness to database systems. The idea then is to extend the same transactional guarantees to new and evolving paradigms, such as Web services. To achieve this, we first need to adapt the transactional mechanisms to the distinguishing characteristics of Web services, mainly composability, long-running nature, and privacy and security concerns. Composability refers to the ability to form new composite services by combining the functionalities of existing services. The existing services may themselves be composite, and this leads to a hierarchical composition. Due to their long-running nature, compensation based mechanisms are usually preferred to provide transactional guarantees for Web services. Compensation consists of semantically undoing the execution effects until failure, and as such requires access (visibility) over the execution details of the services in the composition. However, such visibility may not always be feasible in a compositional context where component services are provided by different providers across organizational boundaries, with very strong privacy and security constraints. This paper looks at compensation options for Web services in a hierarchical composition. A compensation can be done by the provider who executed the service or by a different provider. Multiple compensation options may be available for a composite service both at the same level and at different levels of the hierarchy. Some options may compensate just the execution of an individual service whereas some others may also compensate the executions of some other services invoked by that service. Each compensation option has a cost associated with it. In this paper, we show how to find an optimal compensation option under restricted visibility.
Industry and researchers acknowledge Web services as being the next generation of distributed com... more Industry and researchers acknowledge Web services as being the next generation of distributed computing. However, several issues especially the reliability aspect needs to be addressed before Web services can deliver its promise. Due to their heterogeneous, autonomous and long-lived nature, traditional ACID (Atomicity, Consistency, Integrity, Durability) based models are not sufficient for providing transactional guarantee to Web services compositions. To overcome this limitation, many extended transaction models have been proposed based on the concept of compensation. In this paper, we stress on the importance of two aspects, the Cost of Compensation and End User Involvement, which are missing from most of the transaction models proposed until now. We also show how industry standards like BPEL4WS, WS-Transaction can be augmented to facilitate the above aspects. Finally, we propose a simple classification towards describing compensating operations.
The most promising feature of the Web services platform is its ability to form new services by co... more The most promising feature of the Web services platform is its ability to form new services by combining the capabilities of existing services, i.e., its compos ability. The existing services may themselves be composed of other services, leading to a hierarchical composition. In a hierarchical composition, providers vary in the visibility they have over the other providers in the composition. For example, a provider may not be aware of any providers in the hierarchy other than its parent and children. On the other hand, a provider may be aware of all other providers in the hierarchy. Towards this end, we introduce the notion of spheres of visibility (SoV) as an abstraction to capture the upward/downward visibility aspects of the providers in a hierarchical composition. The latter part of this paper deals with compensation. We outline a compensation mechanism for hierarchical compositions conforming to the visibility restrictions modeled as SoV.
We consider hierarchical systems where nodes represent entities and edges represent binary relati... more We consider hierarchical systems where nodes represent entities and edges represent binary relationships among them. An example is a hierarchical composition of Web services where the nodes denote services and edges represent the parent-child relationship of a service invoking another service. A fundamental issue to address in such systems is, for two nodes X and Y in the hierarchy whether X can see Y, that is, whether X has visibility over Y. In a general setting, X seeing Y may depend on (i) X wishing to see Y, (ii) Y wishing to be seen by X, and (iii) other nodes not objecting to X seeing Y. The visibility could be with respect to certain attributes like operational details, execution logs, security related issues, etc. In this paper, we develop a generic conceptual model to express visibility. We study two complementary notions: sphere of visibility of a node X that includes all the nodes in the hierarchy that X sees; and sphere of noticeability of X that includes all the nodes that see X. We also identify the dual properties, coherence and correlation, that relate the visibility and noticeability notions. We propose elegant methods of constructing the spheres with these properties.
The most promising feature of the Web services platform is its ability to form new services by co... more The most promising feature of the Web services platform is its ability to form new services by combining the capabilities of already existing services, i.e., its composability. The existing services may themselves be composed of other services, leading to a hierarchical composition. In this work, we focus on the monitoring aspect for hierarchical Web services compositions. We are primarily interested in capturing the state of a hierarchical composition at any given point of time (snapshot). We discuss in detail how some of the snapshot algorithms proposed in literature can be extended in a Web services context. Snapshots usually reflect a state of the system which “might have occurred”. Towards this end, we show how we can acquire a state that “actually occurred” from such snapshots. Finally, we discuss the different types of execution related queries and how we can answer them using the captured snapshots.
For complex services, logging is an integral part of many middleware aspects, especially, transac... more For complex services, logging is an integral part of many middleware aspects, especially, transactions and monitoring. In the event of a failure, the log allows us to deduce the cause of failure (diagnosis), recover by compensating the logged actions (atomicity), etc. However, for heterogeneous services, logging all the actions is often impracticable due to privacy/security constraints. Also, logging is expensive in terms of both time and space. Thus, we are interested in determining the absolute minimal number of actions that needs to be logged, to know with certainty the actual sequence of executed actions from any given partial log. This problem happens to be NP-Complete. We propose a decomposition framework in order to use a divide and conquer algorithm. This method dramatically decreases the complexity for hierarchical services (up to 2 exponentials) and can also be used in distributed services.
For complex services, logging is an integral part of many middleware aspects, especially, transac... more For complex services, logging is an integral part of many middleware aspects, especially, transactions and monitoring. In the event of a failure, the log allows us to deduce the cause of failure (diagnosis), recover by compensating the logged actions (atomicity), etc. However, for heterogeneous services, logging all the actions is often impracticable due to privacy/security constraints. Also, logging is expensive in terms of both time and space. Thus, we are interested in determining the absolute minimal number of actions that needs to be logged, to know with certainty the actual sequence of executed actions from any given partial log. This problem happens to be NP-Complete. We consider complex services represented as a hierarchy of services, and propose a decomposition mechanism which dramatically decreases the complexity (up to 2 exponentials). The decomposition also works for distributed services.
For complex services, logging is an integral part of many middleware aspects, especially, transac... more For complex services, logging is an integral part of many middleware aspects, especially, transactions and monitoring. In the event of a failure, the log allows us to deduce the cause of failure (diagnosis), recover by compensating the logged actions (atomicity), etc. However, for heterogeneous services, logging all the actions is often impracticable due to privacy/security constraints. Also, logging is expensive in terms of both time and space. Thus, we are interested in determining a small number of actions that needs to be logged, to know with certainty the actual sequence of executed actions from any given partial log. We propose two heuristics to determine such a small set of transitions, with services modeled as Finite State Machines. The first one is based on (Positive) Discrimination of transitions, using every observation to know (discriminate) that a maximal number of transitions occurred. We characterize it algebraically, giving a very fast algorithm. The second algorithm, the distinguishing algorithm, uses every observation to maximize the number of transitions which are ensured not to have occurred. We show experimentally that the second algorithm gives much more accurate results than the first one, although it is also slower (but still fast enough). * This work is supported by la Region Bretagne (CREATE ACTIVE-DOC) and ANR-06-MDCA-005 DOCFLOW.
Over the years, the notion of transactions has become synonymous with providing fault-tolerance, ... more Over the years, the notion of transactions has become synonymous with providing fault-tolerance, reliability and robustness to database systems. However, challenges arise when we try to apply them to novel computing paradigms such as ActiveXML (AXML) systems. AXML provides an elegant platform to integrate the power of XML, Web services and Peer to Peer (P2P) paradigms by allowing (active) Web services calls to be embedded within XML documents. We propose a transactional framework which provides relaxed ACID properties to AXML systems. Relaxed atomicity is usually provided with the help of compensation. However, current compensation based models assume the existence of a pre-defined compensating operation. Also, compensation is assumed to be more or less peer (or service provider) dependent, i.e., the original and compensating services are provided by the same peer. We show how compensation for AXML transactions can be constructed dynamically at run-time and achieved in a peer independent manner. Finally, we consider the issue of peer disconnection, an inherent trait of P2P systems, and propose an innovative solution based on peer "chaining".
With more and more data stored into XML databases, there is a need to provide the same level of f... more With more and more data stored into XML databases, there is a need to provide the same level of failure resilience and robustness that users have come to expect from relational database systems. In this work, we discuss strategies to provide the transactional aspect of atomicity to XML databases. The main contribution of this paper is to propose a novel approach for performing updates-in-place on XML databases, with the undo statements stored in the same high level language as the update statements. Finally, we give experimental results to study the performance/storage trade-off of the updates-in-place strategy (based on our undo proposal) against the deferred updates strategy to providing atomicity.
... Debmalya Biswas Nokia Research Center, Lausanne, Switzerland debmalya.biswas@ nokia.com ... M... more ... Debmalya Biswas Nokia Research Center, Lausanne, Switzerland debmalya.biswas@ nokia.com ... Many apps (including some of the most popular ones) have been observed to misuse the install-time access given to them at run-time. ...
Secure Multi-party Computation (SMC) enables secure distributed computation of arbitrary function... more Secure Multi-party Computation (SMC) enables secure distributed computation of arbitrary functions of private inputs. Multiple techniques for SMC have been well studied and can be applied within cryptographic protocols, leading to large and complex protocols. Their implementation is difficult for an average programmer to understand, time consuming and potentially prone to errors. We introduce a new programming language dedicated to cryptographic protocols, which speeds up their implementation, the deployment of the running software, and furthermore provides integrated support for benchmarking.
Knowing the preferences and habits of an individual customer, it is possible to offer to that cus... more Knowing the preferences and habits of an individual customer, it is possible to offer to that customer well customized and adapted services, matching his needs and desires. This is advantageous for the entity offering the service (e.g., a retailer) as well, as it helps in creating additional sales or improve customer retention. The main unsolved problem today is that the profile of each individual customer would be necessary in order to create such services, posing severe risks regarding privacy and data protection. This paper proposes efficient encryption schemes that allow profiling to be outsourced while preserving privacy. The schemes ensure that the customer is always in control of his profile data, at the same time making shopping data across multiple retailers available to third party service providers to be able to provide targeted services.
We consider large scale Publish/Subscribe systems deployed across multiple organizations. However... more We consider large scale Publish/Subscribe systems deployed across multiple organizations. However, such cross organizational deployment is often hindered by firewalls and Network Address Translators (NATs). Several workarounds have been proposed to allow firewall and NAT traversal, e.g. VPN, connection reversal, relay routers. However, each traversal mechanism in turn leads to trade-offs with respect to implementation complexity, infrastructure overhead, latency, etc. We focus on the latency aspect in this work. We propose a cost-performance model that allows quantitative evaluation of the performance latency induced by the different firewall traversal mechanisms. The utility of the model is that for a given network configuration, it is able to provide a (close) approximation of the performance latencies based on simulation results, without actually having to deploy them in practice. This also allows selecting the best traversal mechanism for a given configuration. Finally, experimental results are given to show the validity of the proposed model.
Knowing the preferences and habits of an individual customer, it is possible to offer to that cus... more Knowing the preferences and habits of an individual customer, it is possible to offer to that customer well customized and adapted services, matching his needs and desires. This is advantageous for the entity offering the service (e.g., a retailer) as well, as it helps in creating additional sales or improve customer retention. The main unsolved problem today is that the profile of each individual customer would be necessary in order to create such services, posing severe risks regarding privacy and data protection. This paper proposes efficient encryption schemes that allow profiling to be outsourced while preserving privacy. The schemes ensure that the customer is always in control of his profile data, at the same time making shopping data across multiple retailers available to third party service providers to be able to provide targeted services.
Social Network Analysis (SNA) is now a commonly used tool in criminal investigations, but evidenc... more Social Network Analysis (SNA) is now a commonly used tool in criminal investigations, but evidence gathering and analysis is often restricted by data privacy laws. We consider the case where multiple investigators want to collaborate but do not yet have sufficient evidence that justifies a plaintext data exchange. We propose a practical solution that allows an investigator to expand his current view without actually exchanging sensitive private information. The investigator gets a partially anonymized view of the entire social network, while preserving his known view.
Secure Multiparty Computation (SMC) has gained tremendous importance with the growth of the Inter... more Secure Multiparty Computation (SMC) has gained tremendous importance with the growth of the Internet and E-commerce, where mutually untrusted parties need to jointly compute a function of their private inputs. However, SMC protocols usually have very high computational complexities, rendering them practically unusable. In this paper, we tackle the problem of comparing two input values in a secure distributed fashion. We propose efficient secure comparison protocols for both the homomorphic encryption and secret sharing schemes. We also give experimental results to show their practical relevance.
Uploads
Papers by Debmalya Biswas