445 research outputs found
Note sur les globules sanguins du Mermis aquatilis : suivie de quelques remarques sur la structure anatomique de cette espèce
No full textpar Ed. Bugnion(Extrait des Actes de la 60. session de la Soc. Helv. des Sc.Nat.
Métamorphoses du Meigenia bisignata (mouche parasite de la tribu des tachinaires)
No full textpar Edouard BugnionAus: Bulletin de la Société vaudoise des sciences naturelles. - 1884, vol. 17, 188
IX : An OS for datacenter applications with aggressive networking requirements
No full textThe conventional wisdom is that aggressive networking requirements, such as high packet rates for small messages and microsecond-scale tail latency, are best addressed outside the kernel, in a user-level networking stack. We present IX, a dataplane operating system designed to support low-latency, high-throughput and high-connection count applications. Like classic operating systems such as Linux, IX provides strong protection guarantees to the networking stack. However, and unlike classic operating systems, IX is designed for the ground up to support applications with aggressive networking requirements on dense multi-core platforms with 10GbE and 40GbE Ethernet NICs. IX outperforms Linux by an order of magnitude on micro benchmarks, and by up to 3.6x when running an unmodified memcached, a popular key-value store.
The presentation is based on the joint work with Adam Belay, George Prekas, Ana Klimovic, Sam Grossman and Christos Kozyrakis, published at OSDI 2014; Best Paper Award.
About the speaker
Prof. Bugnion joined EPFL in 2012, where his focus is on datacenter systems. His areas of interest include operating systems, datacenter infrastructure (systems and networking), and computer architecture. Before joining EPFL, Edouard spent 18 years in the US, where he studied at Stanford and co-founded two startups: VMware and Nuova Systems (acquired by Cisco). At VMware from 1998 until 2005, he played many roles including CTO. At Nuova/Cisco from 2005 until 2011, he helped build the core engineering team and became the VP/CTO of Cisco's Server, Access, and Virtualization Technology Group, a group that brought to market Cisco's Unified Computing System (UCS) platform for virtualized datacenters. Together with his colleagues, he received the ACM Software System Award for VMware 1.0 in 2009. His paper "Disco: Running Commodity Operating Systems on Scalable Multiprocessors" received a Best Paper Award at SOSP '97 and was entered into the ACM SIGOPS Hall of Fame Award in 2008.</p
Faune des coléoptères du Valais et des régions limitrophes
No full textpar Emile Favre ; avec la collaboration du Dr Edouard BugnionHandschriftliches Geschenkexlibris: "A la bibliothèque du Musée entomologique de Zurich. Hommage de l'auteur" 002190343_0001 Exemplar der ETH-BIBExlibrisstempel: "Entomologische Sammlung der eidgen. techn. Hochschule Zürich" 011145977_0001 Exemplar der ETH-BI
Edouard Roditi Collection 1931-1980
No full textThe collection documents professional activities of Edouard Roditi as an art historian and critic and consists of manuscripts, notes, research files, and a wealth of art catalogues, press release, photographs, and exhibit invitations. There are manuscripts by Roditi as well as by other authors on such topics as Jewish artists in France, Sephardic Jews, and other. Correspondence collected here includes Roditi’s professional correspondence with individuals as well as organizations.However, the bulk of the collection consists of art catalogues, press release, photographs, and exhibit invitations dealing with Jewish artists in France. These materials have not been microfilmed and are available in the original form. There are no personal materials that shed light on Eduard Roditi’s life, nor any materials pertaining to his work as an interpreter for the Nuremberg trial, or San Francisco Conference, during which the United Nations Organization was established.Edouard Roditi was born in France into an American family. He was educated in England, France, Germany, and in the United States. During WWII and shortly after he worked for a number of United States agencies. He was an author as well as a well established art critic and a translator. - Edouard Roditi died on May 10, 1992, at the age of 81.Photographs removed to Photograph Collectiondigitize
Scaling Out Bioinformatics in the Data Center
No full textWhole Genome Sequencing is a process in the field of bioinformatics that transforms biological samples of DNA into an electronic dataset of genetic bases. The process consists of two sequential components. First, the laboratory process of sequencing transforms the biological DNA samples into a digital format by transcribing the sequence of bases that make up short snippets of DNA. Second, a genomic workflow uses software to transform this data into a representation that is useful for genomic analysis. Recent advancements in High-Throughput Sequencing technology enable the laboratory phase to produce data faster and at a lower cost than prior techniques were capable of. The applications and file formats used in workflows have not undergone commensurate technological advancement in order to accommodate this deluge of genomic data.
This thesis introduces a redesign of genomic workflows, their component applications, and the underlying file formats in order to scale out workflows across a data center. The design builds upon on two design components. First, a unified file format supplants the myriad existing formats in order to accommodate scale-out multi-machine I/O. The file format imposes minimal feature requirements upon the storage system, thereby enabling its use in high- performance systems for processing and cost-effective cold storage systems for long-term storage. Second, a new cloud computing framework provides an API for composing workflows in an abstract logical description and delegating the execution of the logic to a common runtime. The framework's runtime executes the logical workflow description on scale-out hardware resources while abstracting the execution details.
We combine the file format and framework to build a new set of workflows that scale out across data center resources. These scale-out workflows incorporate existing workflow applications (compartmentalized into libraries that the framework invokes) and new applications that leverage the features provided by the scale-out architecture. All workflows delegate work distribution and task concurrency to the framework's runtime and utilize a common set of subcomponents for auxiliary code (e.g., I/O with various storage systems, processing different file formats). These workflows are able to scale out across a data center to the point of saturating the throughput of one or more hardware resources.DCS
A Use Case Oriented Survey of Self-Sovereign Identity
No full textWith self-sovereign identity (SSI), we stand at a crossroads that is leading society to a new kind of digital identity. Under this new paradigm, users no longer have to remember a username and a password; instead they gain full power on the information that is issued to them by trusted entities. From their perspective, using SSI therefore consists in carrying a digital wallet, typically in the form of a smartphone application, where so-called verifiable credentials can be stored in a secure environment behind biometrical identifi- cation. However, this high user responsibility raises the first challenge of SSI: wallets are vulnerable and people are prone to losing access to them for various reasons. Therefore, data should be properly backed up, so that credentials can be recovered without altering user experience. In practical terms, having such a new model for digital identity enables many potential use cases, either by digitalizing existing processes from the physical world or by creating new possibilities thanks to the power of combining data. Consequently, it becomes hard to decide which use cases should be prioritized, let alone which of them are well-aligned with the principles of SSI. Therefore, this work establishes a set of requirements, according to which potential use cases of SSI can be assessed qualitatively. Additionally, a threat model for SSI systems is presented, identifying possible attacks on such use cases in order to help developers take design decisions about their implementations. Finally, this work focuses on an educational use case and explores a proof of concept implementation based on hyperledger Indy, which is a decentralized network specifically built for identity purposes; and ACA-Py, which is a framework that provides an interface to communicate with Indy underneath. In this context, there are also a couple of issues related to the versioning of functionalities, the beta status of wallet applications and credential schema formats. This highlights the fact that SSI is not yet ready for a public rollout, but it also motivates research and development to put a strong emphasis on it in the coming few years.DCS
Enhancing Quality of Service metrics for high fan-in Node.js applications by optimising the network stack: Leveraging IX: The Dataplane Operating System
No full textThis thesis investigates the feasibility of porting Node.js, a JavaScript web application framework and server, to IX, a dataplane operating system specifically developed to meet the needs of high performance microsecond-computing type of applications in a datacentre setting. We show that porting requires extensions to the IX kernel to support UDS polling, which we implement. We develop a distributed load generator to benchmark the framework. The results show that running Node.js on IX improves throughput by up to 20.6\%, latency by up to 5.23×, and tail latency by up to 5.68× compared to a Linux baseline. We show how server side request level reordering affect the latency distribution, predominantly in cases where the server is load saturated. Finally, due to various limitations of IX, we are unable at this time to recommend running Node.js on IX in a production environment, despite improved metrics in all test cases. However, the limitations are not fundamental, and could be resolved in future work.DCS
Hidden Filesystem Design and Improvement
No full textPlausible deniability is a strong cryptographic security property which, in the context of data storage, offers protection against invasive and coercive adversaries who have the power to extort the passwords to encrypted data. Concretely, to defend against such threats, the storage must be formatted in such a way that multiple passwords can be used to access it, some disclosing innocent data, others unlocking secret contents: the user can then only reveal passwords from the first set to the adversary, and plausibly deny that more exist. This problem has collected the interest of the computer security community for some time now. Several schemes have been devised over the last two decades, offering various levels of performance and security. Some of them, achieving very good performance, arguably do not provide satisfactory guarantees of deniability. Others, accomplishing full, bulletproof security, suffer instead from severe performance hits (especially in terms of I/O overhead and disk space utilisation). We propose a novel design, a scheme called Shufflecake, which targets a more balanced compromise between performance and security. The level of deniability it offers, while not protecting against attacks in the most stringent threat model, is sufficient in many practical scenarios. On the other hand, it achieves a 99.6% disk efficiency, and a 1x-3x slowdown over regular disk encryption tools, which makes it suited for real-world applications.DCS
Duplicate detection against data leakage in a company setting
No full textThe DupeLeak8 tool provides a standalone and easy to deploy solution for detecting potential data
leaks due to near-duplicates in a large file system. It performs efficient all-to-all file comparisons
which results in a global coverage of the system’s textual duplication issues and compares the
detected duplicate’s classification levels to check for inconsistencies: DupeLeak8 assumes files with
similar content are more likely to have similar classification levels. It also provides users with a one-
to-all comparison method which can be used to monitor duplicated content in environments with
constant file turnover, such as a company file system. DupeLeak8 offers flexible and customizable
settings to adapt to the user’s needs as best as possible.
DupeLeak8 focuses on detecting data breaches due to exact or partial textual duplicates at a
granularity level chosen by the user, by leveraging advantages of both local and global similarity
analysis techniques.
The tool is implemented in Java and relies on widely audited and tested open-source libraries.
DupeLeak8 boasts a user-friendly API, allowing users to execute our application with as few as three
lines of code.
We evaluate DupeLeak8 on a large dataset of 6k enterprise internal documents for a total size of
22GB. All-to-all comparisons took 13 minutes and one-to-all queries can be performed interactively
in less than 1 second.DCS
- …
