公开(公告)号：WO2011009652A3

公开(公告)日：2011-04-14

申请号：PCT/EP2010056727

申请日：2010-05-17

Applicant: IBM ,  IBM FRANCE ,  DELLA CORTE GIANLUCA ,  BORGHETTI STEFANO ,  SGRO' ANTONIO ,  GIANFAGNA LEONIDA

Inventor： DELLA CORTE GIANLUCA ,  BORGHETTI STEFANO ,  SGRO' ANTONIO ,  GIANFAGNA LEONIDA

IPC: G06F9/50

CPC classification number: G06F9/505 ,  G06F9/5072

Abstract: The method of the present invention provides an automatic and optimised selection of the network topology for distributing scheduling of jobs on the computers of the modified network topology. The automatic and optimised selection of the network topology starts from the current topology and a desired number of additional connections. In this way the method of the present invention provides a higher convergence speed for the modified consensus algorithm in comparison e.g. to a simple ring network. The method exploits the so called small-world networks. Small-world networks are more robust to perturbations than other network architectures. The preferred embodiment provides a workload scheduling system which is highly scalable to accommodate increasing workloads within a heterogeneous distributed computing environment. A modified average consensus algorithm is used to distribute network traffic and jobs amongst a plurality of computers.

Abstract translation: 本发明的方法提供了用于在修改的网络拓扑的计算机上分发作业调度的网络拓扑的自动优化选择。 网络拓扑的自动优化选择从当前的拓扑结构和所需数量的附加连接开始。 以这种方式，本发明的方法为修改的共识算法提供了较高的收敛速度。 到一个简单的环网。 该方法利用了所谓的小世界网络。 与其他网络架构相比，小世界网络的干扰更加强大。 优选实施例提供了一种工作量调度系统，其高度可扩展以适应异构分布式计算环境内的增加的工作负载。 改进的平均一致性算法用于在多个计算机之间分配网络流量和作业。

公开(公告)号：WO2014086656A3

公开(公告)日：2014-12-18

申请号：PCT/EP2013074956

申请日：2013-11-28

Applicant: IBM ,  IBM DEUTSCHLAND

Inventor： LANNI LEONARDO ,  TORTOSA ANDREA ,  SCOTTI ALESSANDRO ,  LAZZARO LUCA ,  SGRO' ANTONIO

IPC: G06F9/50 ,  H04L29/08

CPC classification number: H04L67/42 ,  G06F9/5044 ,  G06F9/505 ,  G06F2209/5017 ,  G06F2209/509 ,  H04L43/08 ,  H04L67/02 ,  H04L67/04 ,  H04L67/10 ,  H04L67/22 ,  H04L67/303 ,  H04L67/32 ,  H04L67/34

Abstract: A solution is proposed for controlling a distributed application. A corresponding method (400) comprises the steps of detecting (404-422) an indication of a computational capability of a client computing machine and/or of a server computing machine, building (424-440) at least part of a page in response to a request received by the server computing machine from the client computing machine, the page comprising a set of commands each one for submitting a next request on the client computing machine, a processing logic for serving each next request being split between a client processing component and/or a server processing component, returning (442-456) the page with the client processing component for each command to the client computing machine for causing the client computing machine to load the page and execute each client processing component in response to the corresponding next request, and executing (458-464,424-440) each server processing component in response to the corresponding next request.

Abstract translation: 提出了一种解决方案来控制分布式应用程序。 相应的方法（400）包括以下步骤：检测（404-422）客户端计算机器和/或服务器计算机器的计算能力的指示;响应于建立（424-440）页面的至少一部分 服务器计算机器从客户机计算机器接收到的请求，所述页面包括一组命令，每一个命令用于在客户机计算机器上提交下一个请求;处理逻辑，用于服务每个下一个请求被分割在客户机处理组件 和/或服务器处理组件，向客户端计算机器返回（442-456）具有用于每个命令的客户端处理组件的页面，以使得客户端计算机器加载页面并且响应于相应的响应执行每个客户端处理组件 下一个请求，并且响应于相应的下一个请求执行（458-464,424-440）每个服务器处理组件。

公开(公告)号：WO2010076088A3

公开(公告)日：2010-10-14

申请号：PCT/EP2009065736

申请日：2009-11-24

Applicant: IBM ,  IBM FRANCE ,  BORGHETTI STEFANO ,  SGRO' ANTONIO ,  GIANFAGNA LEONIDA ,  DELLA CORTE GIANLUCA ,  HAAG ALESSANDRO

Inventor： BORGHETTI STEFANO ,  SGRO' ANTONIO ,  GIANFAGNA LEONIDA ,  DELLA CORTE GIANLUCA ,  HAAG ALESSANDRO

IPC: H04L29/06

CPC classification number: H04L63/08 ,  H04L63/10

Abstract: A mechanism is provided for performing secure system access by a requesting user without sharing a password of a credential owner. A database stores system information for resources. The owner of super user authority for a resource provides system information to the database including a credential for accessing the resource. When a user wishes to access the system, client software of the requestor sends an access request to client software of the owner. The client software of the owner prompts the owner to authorize or deny access. Responsive to the owner authorizing the access, the client software of the owner returns authorization to the client software of the requestor, which then uses the credential in the system information database to access the resource. The client software of the requestor does not cache or store the credential or present the credential to the user.

Abstract translation: 提供了一种用于由请求用户执行安全系统访问而不共享凭证所有者的密码的机制。 数据库存储资源的系统信息。 资源的超级用户权限的所有者向数据库提供系统信息，包括用于访问资源的凭证。 当用户希望访问系统时，请求者的客户端软件向访问者的客户端软件发送访问请求。 所有者的客户端软件提示所有者授权或拒绝访问。 响应所有者授权访问，所有者的客户端软件向请求者的客户端软件返回授权，然后请求者使用系统信息数据库中的凭证来访问资源。 请求者的客户端软件不缓存或存储凭证或将凭证呈现给用户。

公开(公告)号：DE112010003027T5

公开(公告)日：2012-08-02

申请号：DE112010003027

申请日：2010-05-17

Applicant: IBM

Inventor： DELLA CORTE GIANLUCA ,  SGRO' ANTONIO ,  GIANFAGNA LEONIDA ,  BORGHETTI STEFANO

IPC: G06F9/50

Abstract: Das Verfahren der vorliegenden Erfindung stellt eine automatische und optimierte Auswahl der Netztopologie für die verteilte Steuerung von Jobs auf den Computern der modifizierten Netztopologie bereit. Die automatische und optimierte Auswahl der Netztopologie geht von der aktuellen Topologie und einer gewünschten Anzahl zusätzlicher Verbindungen aus. Auf diese Weise bietet das Verfahren der vorliegenden Erfindung im Vergleich beispielsweise zu einem einfachen Ringnetz eine höhere Konvergenzgeschwindigkeit für den modifizierten Konsensalgorithmus. Das Verfahren bedient sich der so genannten Kleine-Welt-Netzwerke. Kleine-Welt-Netzwerke sind wesentlich störfester als andere Netzarchitekturen. Die bevorzugte Ausführungsart stellt ein Laststeuersystem bereit, das in hohem Maße skalierbar ist und innerhalb einer heterogenen verteilten Rechnerumgebung wachsende Arbeitsbelastungen bewältigt. Es findet ein modifizierter Durchschnittskonsensalgorithmus Anwendung, um Netzwerkverkehr und Jobs auf eine Vielzahl von Computern zu verteilen.

公开(公告)号：GB2512847A

公开(公告)日：2014-10-15

申请号：GB201306360

申请日：2013-04-09

Applicant: IBM

Inventor： CORTE GIANLUCA DELLA ,  SGRO' ANTONIO ,  DONATELLI ALEX ,  PROIETTI STEFANO

IPC: G06F9/50 ,  G06Q10/06

Abstract: Disclosed is a computer-implemented method of predicting the required number of servers for a future computing workload. The method comprises determining the number of servers in three classes of servers and predicting the future number of servers in the different classes based on a Susceptible-Infected-Recovered-Algorithm. The three classes are, class S are servers that are adapted to deploy the application, class I are servers that deploy the application and class R are servers that are disconnected from the network. The algorithm is based on the deployment rate, un-deployment rate and a removal rate. The deployment rate is the number of servers leaving class S moving to class I, the un-deployment rate is the number of servers leaving class I moving to class S, and the removal rate is the number of servers leaving class I and moving to class R, all in unit time. Such that the total number of servers required in the topology of the computer network in the data centre at a future time t is calculated.