Skip to main content

Paulo Guedes

Universidade Lusófona, Escola de Comunicação, Artes e Tecnologias de Informação, Faculty Member

Followers

22

Following

3

Co-author

1

Public Views

Universidade de Lisboa

Rutgers, The State University of New Jersey

Seoul National University

University of Asia and the Pacific

University of Web Marketing

Google

FLACSO-Ecuador

Luciana Arantes

Interests

Uploads

Papers by Paulo Guedes

Abstract Lightweight Logging for Lazy Release Consistent Distributed Shared Memory

This paper presents a new logging and recovery algorithm for lazy release consistent distributed ... more This paper presents a new logging and recovery algorithm for lazy release consistent distributed shared memory (DSM). The new algorithm tolerates single node failures by maintaining a distributed log of data dependencies in the volatile memory of processes. The algorithm adds very little overhead to the memory consistency protocol: it sends no additional messages during failure-free periods; it adds only a minimal amount of data to one of the DSM protocol messages; it introduces no forced rollbacks of non-faulty processes; and it performs no communication-induced accesses to stable storage. Furthermore, the algorithm logs only a very small amount of data, because it uses the log of memory accesses already maintained by the memory consistency protocol. The algorithm was implemented in TreadMarks, a state-of-the-art DSM system. Experimental results show that the algorithm has near zero time overhead and very low space overhead during failure-free execution, thus refuting the common be...

Distribution and Persistence in the IK Platform: Overview and Evaluation

IK is an object-oriented platform that simplifies the construction of applications that handle pe... more IK is an object-oriented platform that simplifies the construction of applications that handle persistent and distributed data. A single programming paradigm is used to manipulate volatile, persistent and distributed objects uniformly. Object invocation is the basic primitive of the system, embodying all the features required for transparent handling of persistence and distribution. Object references can be freely passed as parameters in remote invocations or stored persistently. Objects are initially created as volatile and are automatically recycled when they are no longer referenced. Objects are promoted to persistent when they become reachable from a persistent root. Persistent objects are read from disk on demand when they are invoked and are written back to disk when the application terminates. Applications are written in a language similar to C++, but with some semantic differences. Classes are represented as special persistent objects. They are dynamically linked when needed...

Security Policy Consistency

With the advent of wide security platforms able to express simultaneously all the policies compri... more With the advent of wide security platforms able to express simultaneously all the policies comprising an organization's global security policy, the problem of inconsistencies within security policies become harder and more relevant. We have defined a tool based on the CHR language which is able to detect several types of inconsistencies within and between security policies and other specifications, namely workflow specifications. Although the problem of security conflicts has been addressed by several authors, to our knowledge none has addressed the general problem of security inconsistencies, on its several definitions and target specifications. 1 Introduction Over the years several access control policies have been proposed in the literature. Although these policies cover many di#erent situations and types of information, they are often considered in isolation. Thus, they are not suitable for organizations with complex structures, that manage simultaneously several types of i...

A checkpoint protocol for an entry consistent shared memory system

Workstation clusters are becoming an interesting alternative to dedicated multiprocessors. In thi... more Workstation clusters are becoming an interesting alternative to dedicated multiprocessors. In this environment, the probability of a failure, during an application’s execution, increases with the execution time and the number of workstations used. If no provision is made for handling failures, it is unlikely that long running applications will terminate successfully. One solution to this problem is process checkpointing. This paper presents a checkpoint protocol for a multithreaded distributed shared memory system based on the entry consistency memory model. The protocol allows transparent recovery from single node failures and, in some cases, from multiple node failures. A simple mechanism is used to determine if the system can be brought to a consistent state in the event of multiple machine crashes. The protocol keeps a distributed log of shared data accesses in the volatile memory of the processes, taking advantage of the independent failure characteristics of workstation cluste...

Verifying Workflow Processes against Organization Security Policies

this paper is to show how a security specification written in a generic and flexible security lan... more

Ship Hull Hydrodynamic Analysis using Distributed Shared Memory

Abstract. In this article we present the results obtained from the execution of a commercial Comp... more Abstract. In this article we present the results obtained from the execution of a commercial Computational Fluid Dynamics program on a cluster of personal computers. The communication and data interchange between the nodes uses a Distributed Shared Memory (DSM) system called TreadMarks. The parallelized program was run on a 100Mbit/s Ethernet network connecting 4 personal computers. Using data similar to a real execution on a industrial environment, we achieved a maximum speedup of about 1.7 and measured an allmost constant communication time with various number of nodes. Comparing our DSM solution with a similar parallelization approach but using Message passing we observed that with less effort we can accomplish better performance with less communication if using a DSM system. 1

Efficient Error-Propagating Block Chaining

This document presents EPBC, Efficient Error-Propagating Block Chaining, a new and efficient bloc... more This document presents EPBC, Efficient Error-Propagating Block Chaining, a new and efficient block encryption mode using both plaintext and ciphertext feedback. This encryption mode is similar to another one, IOBC, and was likewise designed to propagate erroneous decryptions of tampered blocks of ciphered data to all following blocks, hence allowing to validate the integrity of that data using a predefined trailing value. However, EPBC is more secure than IOBC, as it is not vulnerable to any known-plaintext attacks, and is more efficient than IOBC. Performance tests ran on a SPARCstation 10/40 show that EPBC is in average 1.2 times faster than IOBC, and 6.3 to 10.9 times faster than a common combination of an encryption mode and a one-way hash function (CBC and MD5).

Object-Oriented Interfaces in the Mach 3.0 Multi-Server System

The Mach 3.0 multi-server system decomposes the functionality of the operating system between a m... more The Mach 3.0 multi-server system decomposes the functionality of the operating system between a micro-kernel, a set of system servers running in user-mode and an emulation library executing in the address space of applications. The interfaces provided by the system servers are objectoriented and both the servers and the emulation library are written in an object-oriented language. In this paper we present how the interfaces between the components are specified and implemented to guarantee consistency and early detection of errors, yet maintaining the flexibility to extend and configure the system by adding new or modified servers without affecting existing pieces. 1 Introduction A current trend in operating system design is to attempt to distribute the complete system functionality between a micro-kernel and a set of system servers executing in user mode and communicating by message exchange [6, 1]. This approach provides a clean separation between the different parts of the system,...

Design of the PerDiS Transactional File System

We present the design for a fault-tolerant persistent data storage in a wide-area distributed coo... more We present the design for a fault-tolerant persistent data storage in a wide-area distributed cooperative engineering environment. We discuss some of the current options in file system design and, in particular, the relation between file mobility and transaction protocols. We believe that file mobility, and consequently coherence protocols, global file naming and versioning, are issues whose exploration and use will lead to improved transaction protocols and to an increased applicability of transactional file systems. Our aim is to implement a wide area distributed and fault tolerant file system in the context of the current PerDiS (Persistent Distributed Store) ESPRIT project[SHA97]. The emphasis of our design concerns is placed on improving the transactional and fault tolerant behavior of distributed file systems.

PerDiS: design, implementation, and use of a PERsistent DIstributed Store

Parallel Operating Systems

ion Layer (HAL) - All these components are layered on top of a hardware abstraction layer. This l... more ion Layer (HAL) - All these components are layered on top of a hardware abstraction layer. This layer hides hardware specific details, such as I/O interfaces and interrupt controllers, from the NT executive thereby enhancing its portability. For example, all cache coherency and flushing in a SMP is hidden beneath this level. The modules that are executed in user mode are: -- System & Service Processes - These operating system processes provide several services such as session control, logon management, RPC and event logging. -- User Applications -- Environment Subsystems - The environment subsystems use the generic operating system services provided by the executive to give applications the interface of a particular operating system. NT includes environment subsystems for MS-DOS, 16 bit Windows, POSIX, OS/2 and for its main environment, the Win32 API. In NT, operating system resources are represented by objects. Kernel objects include processes and threads, file system components (f...

SPL: An access control language for security policies with complex constraints

Most organizations use several security policies to control different systems and data, comprisin... more Most organizations use several security policies to control different systems and data, comprising in this way a global complex policy. These security policies are often scattered over different environments, each one with its own security model and domain of administration, making them difficult to administer and understand. Moreover, some applications (e.g. workflow), often need to cross several of these security domains and satisfy each one of their policies, which is very difficult to accomplish when these policies are scattered over the organization, in conflict with each other and frequently expressed in different models. This work presents a security policy language that allows organizations to express and keep their global security policies in one single description. Although flexible enough to express simultaneously several types of complex security policies, this language can be easily implemented by an event monitor. The proposed language can express the concepts of permi...

Transparent Authentication and Confidentiality for Stream Sockets

PES, Privacy Enhanced Sockets, is a user-level subsystem providing network privacy for stream soc... more PES, Privacy Enhanced Sockets, is a user-level subsystem providing network privacy for stream sockets. This document presents its design and implementation in UNIX and Microsoft Windows systems. PES provides secure channels on top of standard stream sockets, allowing existing client-server applications to transparently use secure channels instead of standard remote connections. PES is composed of two entities which can be almost transparently added to existing systems: a dynamic library and a trusted proxy server. All cryptographic details concerning the establishment of secure channels are solely handled by instances of these two entities using the Diffie-Hellman public-key algorithm. Man-in-the-Middle attacks are prevented using asymmetric cryptography for host authentication. Host’s public keys are acquired and stored on a per-user basis such as for PGP, which does not require central management but only a reduced user intervention. PES has been tested with several applications –...

Transparent Authentication and Confidentiality for Datagram Sockets

PES, Privacy Enhanced Sockets, is a user level subsystem providing network confidentiality and ho... more PES, Privacy Enhanced Sockets, is a user level subsystem providing network confidentiality and host authentication forexisting client-serverapplicationsusing socket-based communications. PES was initially designed to provide secure channels on top of connectionoriented protocols, namely TCP/IP, in UNIX and Microsoft Windows systems. This document presents the evolution of PES in order to provide the same security attributes on top of connectionless protocols, namely UDP.

Ik-2 implementation report

Two-Level Recovery For Distributed Shared Memory

Design of the PerDiS Transactional File System

We present the design for a fault-tolerant persistent data storage in a wide-area distributed coo... more We present the design for a fault-tolerant persistent data storage in a wide-area distributed cooperative engineering environment. We discuss some of the current options in file system design and, in particular, the relation between file mobility and transaction protocols. We believe that file mobility, and consequently coherence protocols, global file naming and versioning, are issues whose exploration and use will lead to improved transaction protocols and to an increased applicability of transactional file systems.

The distributed operating system of the SMD project

by Nuno Guimarães and Paulo Guedes

Software: Practice …, Jan 1, 1988

Verifying workflow processes against organization security policies

Proceedings. IEEE 8th International Workshops on Enabling Technologies: Infrastructure for Collaborative Enterprises (WET ICE'99), 1999

Workflow applications for large complex organizations often need to cross several security domain... more Workflow applications for large complex organizations often need to cross several security domains, each with different management and specific security requirements. The resultant cross-dependency between the workflow specification and the security policy of each domain can be hard to manage without specific tools.

Operating system support for an object-oriented environment

Proceedings of the Second Workshop on Workstation Operating Systems, 1989

An overview is given of the Comandos project of the European Strategic Programme for Research on ... more An overview is given of the Comandos project of the European Strategic Programme for Research on Information Technology (ESPRIT). The initial phase of the project was devoted to the definition of the Comandos virtual machine and resulted from contributions of the different partners in the project. Then several testbed implementations were started to validate the basic ideas of the architecture.

Abstract Lightweight Logging for Lazy Release Consistent Distributed Shared Memory

This paper presents a new logging and recovery algorithm for lazy release consistent distributed ... more This paper presents a new logging and recovery algorithm for lazy release consistent distributed shared memory (DSM). The new algorithm tolerates single node failures by maintaining a distributed log of data dependencies in the volatile memory of processes. The algorithm adds very little overhead to the memory consistency protocol: it sends no additional messages during failure-free periods; it adds only a minimal amount of data to one of the DSM protocol messages; it introduces no forced rollbacks of non-faulty processes; and it performs no communication-induced accesses to stable storage. Furthermore, the algorithm logs only a very small amount of data, because it uses the log of memory accesses already maintained by the memory consistency protocol. The algorithm was implemented in TreadMarks, a state-of-the-art DSM system. Experimental results show that the algorithm has near zero time overhead and very low space overhead during failure-free execution, thus refuting the common be...

Distribution and Persistence in the IK Platform: Overview and Evaluation

IK is an object-oriented platform that simplifies the construction of applications that handle pe... more IK is an object-oriented platform that simplifies the construction of applications that handle persistent and distributed data. A single programming paradigm is used to manipulate volatile, persistent and distributed objects uniformly. Object invocation is the basic primitive of the system, embodying all the features required for transparent handling of persistence and distribution. Object references can be freely passed as parameters in remote invocations or stored persistently. Objects are initially created as volatile and are automatically recycled when they are no longer referenced. Objects are promoted to persistent when they become reachable from a persistent root. Persistent objects are read from disk on demand when they are invoked and are written back to disk when the application terminates. Applications are written in a language similar to C++, but with some semantic differences. Classes are represented as special persistent objects. They are dynamically linked when needed...

Security Policy Consistency

With the advent of wide security platforms able to express simultaneously all the policies compri... more With the advent of wide security platforms able to express simultaneously all the policies comprising an organization's global security policy, the problem of inconsistencies within security policies become harder and more relevant. We have defined a tool based on the CHR language which is able to detect several types of inconsistencies within and between security policies and other specifications, namely workflow specifications. Although the problem of security conflicts has been addressed by several authors, to our knowledge none has addressed the general problem of security inconsistencies, on its several definitions and target specifications. 1 Introduction Over the years several access control policies have been proposed in the literature. Although these policies cover many di#erent situations and types of information, they are often considered in isolation. Thus, they are not suitable for organizations with complex structures, that manage simultaneously several types of i...

A checkpoint protocol for an entry consistent shared memory system

Workstation clusters are becoming an interesting alternative to dedicated multiprocessors. In thi... more Workstation clusters are becoming an interesting alternative to dedicated multiprocessors. In this environment, the probability of a failure, during an application’s execution, increases with the execution time and the number of workstations used. If no provision is made for handling failures, it is unlikely that long running applications will terminate successfully. One solution to this problem is process checkpointing. This paper presents a checkpoint protocol for a multithreaded distributed shared memory system based on the entry consistency memory model. The protocol allows transparent recovery from single node failures and, in some cases, from multiple node failures. A simple mechanism is used to determine if the system can be brought to a consistent state in the event of multiple machine crashes. The protocol keeps a distributed log of shared data accesses in the volatile memory of the processes, taking advantage of the independent failure characteristics of workstation cluste...

Verifying Workflow Processes against Organization Security Policies

this paper is to show how a security specification written in a generic and flexible security lan... more

Ship Hull Hydrodynamic Analysis using Distributed Shared Memory

Abstract. In this article we present the results obtained from the execution of a commercial Comp... more Abstract. In this article we present the results obtained from the execution of a commercial Computational Fluid Dynamics program on a cluster of personal computers. The communication and data interchange between the nodes uses a Distributed Shared Memory (DSM) system called TreadMarks. The parallelized program was run on a 100Mbit/s Ethernet network connecting 4 personal computers. Using data similar to a real execution on a industrial environment, we achieved a maximum speedup of about 1.7 and measured an allmost constant communication time with various number of nodes. Comparing our DSM solution with a similar parallelization approach but using Message passing we observed that with less effort we can accomplish better performance with less communication if using a DSM system. 1

Efficient Error-Propagating Block Chaining

This document presents EPBC, Efficient Error-Propagating Block Chaining, a new and efficient bloc... more This document presents EPBC, Efficient Error-Propagating Block Chaining, a new and efficient block encryption mode using both plaintext and ciphertext feedback. This encryption mode is similar to another one, IOBC, and was likewise designed to propagate erroneous decryptions of tampered blocks of ciphered data to all following blocks, hence allowing to validate the integrity of that data using a predefined trailing value. However, EPBC is more secure than IOBC, as it is not vulnerable to any known-plaintext attacks, and is more efficient than IOBC. Performance tests ran on a SPARCstation 10/40 show that EPBC is in average 1.2 times faster than IOBC, and 6.3 to 10.9 times faster than a common combination of an encryption mode and a one-way hash function (CBC and MD5).

Object-Oriented Interfaces in the Mach 3.0 Multi-Server System

The Mach 3.0 multi-server system decomposes the functionality of the operating system between a m... more The Mach 3.0 multi-server system decomposes the functionality of the operating system between a micro-kernel, a set of system servers running in user-mode and an emulation library executing in the address space of applications. The interfaces provided by the system servers are objectoriented and both the servers and the emulation library are written in an object-oriented language. In this paper we present how the interfaces between the components are specified and implemented to guarantee consistency and early detection of errors, yet maintaining the flexibility to extend and configure the system by adding new or modified servers without affecting existing pieces. 1 Introduction A current trend in operating system design is to attempt to distribute the complete system functionality between a micro-kernel and a set of system servers executing in user mode and communicating by message exchange [6, 1]. This approach provides a clean separation between the different parts of the system,...

Design of the PerDiS Transactional File System

We present the design for a fault-tolerant persistent data storage in a wide-area distributed coo... more We present the design for a fault-tolerant persistent data storage in a wide-area distributed cooperative engineering environment. We discuss some of the current options in file system design and, in particular, the relation between file mobility and transaction protocols. We believe that file mobility, and consequently coherence protocols, global file naming and versioning, are issues whose exploration and use will lead to improved transaction protocols and to an increased applicability of transactional file systems. Our aim is to implement a wide area distributed and fault tolerant file system in the context of the current PerDiS (Persistent Distributed Store) ESPRIT project[SHA97]. The emphasis of our design concerns is placed on improving the transactional and fault tolerant behavior of distributed file systems.

PerDiS: design, implementation, and use of a PERsistent DIstributed Store

Parallel Operating Systems

ion Layer (HAL) - All these components are layered on top of a hardware abstraction layer. This l... more ion Layer (HAL) - All these components are layered on top of a hardware abstraction layer. This layer hides hardware specific details, such as I/O interfaces and interrupt controllers, from the NT executive thereby enhancing its portability. For example, all cache coherency and flushing in a SMP is hidden beneath this level. The modules that are executed in user mode are: -- System & Service Processes - These operating system processes provide several services such as session control, logon management, RPC and event logging. -- User Applications -- Environment Subsystems - The environment subsystems use the generic operating system services provided by the executive to give applications the interface of a particular operating system. NT includes environment subsystems for MS-DOS, 16 bit Windows, POSIX, OS/2 and for its main environment, the Win32 API. In NT, operating system resources are represented by objects. Kernel objects include processes and threads, file system components (f...

SPL: An access control language for security policies with complex constraints

Most organizations use several security policies to control different systems and data, comprisin... more Most organizations use several security policies to control different systems and data, comprising in this way a global complex policy. These security policies are often scattered over different environments, each one with its own security model and domain of administration, making them difficult to administer and understand. Moreover, some applications (e.g. workflow), often need to cross several of these security domains and satisfy each one of their policies, which is very difficult to accomplish when these policies are scattered over the organization, in conflict with each other and frequently expressed in different models. This work presents a security policy language that allows organizations to express and keep their global security policies in one single description. Although flexible enough to express simultaneously several types of complex security policies, this language can be easily implemented by an event monitor. The proposed language can express the concepts of permi...

Transparent Authentication and Confidentiality for Stream Sockets

PES, Privacy Enhanced Sockets, is a user-level subsystem providing network privacy for stream soc... more PES, Privacy Enhanced Sockets, is a user-level subsystem providing network privacy for stream sockets. This document presents its design and implementation in UNIX and Microsoft Windows systems. PES provides secure channels on top of standard stream sockets, allowing existing client-server applications to transparently use secure channels instead of standard remote connections. PES is composed of two entities which can be almost transparently added to existing systems: a dynamic library and a trusted proxy server. All cryptographic details concerning the establishment of secure channels are solely handled by instances of these two entities using the Diffie-Hellman public-key algorithm. Man-in-the-Middle attacks are prevented using asymmetric cryptography for host authentication. Host’s public keys are acquired and stored on a per-user basis such as for PGP, which does not require central management but only a reduced user intervention. PES has been tested with several applications –...

Transparent Authentication and Confidentiality for Datagram Sockets

PES, Privacy Enhanced Sockets, is a user level subsystem providing network confidentiality and ho... more PES, Privacy Enhanced Sockets, is a user level subsystem providing network confidentiality and host authentication forexisting client-serverapplicationsusing socket-based communications. PES was initially designed to provide secure channels on top of connectionoriented protocols, namely TCP/IP, in UNIX and Microsoft Windows systems. This document presents the evolution of PES in order to provide the same security attributes on top of connectionless protocols, namely UDP.

Ik-2 implementation report

Two-Level Recovery For Distributed Shared Memory

Design of the PerDiS Transactional File System

We present the design for a fault-tolerant persistent data storage in a wide-area distributed coo... more We present the design for a fault-tolerant persistent data storage in a wide-area distributed cooperative engineering environment. We discuss some of the current options in file system design and, in particular, the relation between file mobility and transaction protocols. We believe that file mobility, and consequently coherence protocols, global file naming and versioning, are issues whose exploration and use will lead to improved transaction protocols and to an increased applicability of transactional file systems.

The distributed operating system of the SMD project

by Nuno Guimarães and Paulo Guedes

Software: Practice …, Jan 1, 1988

Verifying workflow processes against organization security policies

Proceedings. IEEE 8th International Workshops on Enabling Technologies: Infrastructure for Collaborative Enterprises (WET ICE'99), 1999

Workflow applications for large complex organizations often need to cross several security domain... more Workflow applications for large complex organizations often need to cross several security domains, each with different management and specific security requirements. The resultant cross-dependency between the workflow specification and the security policy of each domain can be hard to manage without specific tools.

Operating system support for an object-oriented environment

Proceedings of the Second Workshop on Workstation Operating Systems, 1989

An overview is given of the Comandos project of the European Strategic Programme for Research on ... more An overview is given of the Comandos project of the European Strategic Programme for Research on Information Technology (ESPRIT). The initial phase of the project was devoted to the definition of the Comandos virtual machine and resulted from contributions of the different partners in the project. Then several testbed implementations were started to validate the basic ideas of the architecture.