-
公开(公告)号：GB2497496B
公开(公告)日：2015-01-14
申请号：GB201306367
申请日：2011-08-31
Applicant: IBM
Inventor： O'CONNELL BRIAN MARSHALL , AMSTERDAM JEFFERY DAVID , JENSEN BRIAN WILLIS , WALKER KEITH RAYMOND
Abstract: A computer implemented method, system, and computer program product for managing client connections to a server among a plurality of clients. A messaging server detects, by a first server, an attribute of a client connection that exceeds a threshold, wherein the attribute is a select write operation. The messaging server selects a client that connects to a server, responsive to detecting the attribute of the client exceeding the threshold of select write calls. The messaging server selects a connect mechanism of a second server, responsive to selecting the client. The messaging server transmits a migrate message to the client, the migrate message comprising at least a uniform resource identifier (URI).
-
公开(公告)号：GB2500536A
公开(公告)日：2013-09-25
申请号：GB201311740
申请日：2011-12-13
Applicant: IBM
Inventor： O'CONNELL BRIAN MARSHALL , WALKER KEITH RAYMOND
IPC: H04L47/762
Abstract: Embodiments for server connection optimization in a persistently connected publish/subscribe environment are provided. A computer assigns an initial activity category for a client. The computer logs, into a log file, one or more select operations of a connection between a client and a first server. The computer calculates a ratio of a number of the select operations to a check time of the first server, wherein the check time is a period of time that has elapsed since a previous review of logged data. The computer sets a second activity category for the client based on a look-up of the ratio in a schedule of ratio ranges in a control file, wherein the second activity category is not the initial activity category. The computer transfers the client from the first server to a second server, in response to setting the second activity category.
-
公开(公告)号：GB2501656B
公开(公告)日：2014-08-20
申请号：GB201314915
申请日：2012-02-22
Applicant: IBM
Inventor： JONES ANDREW R , O'CONNELL BRIAN MARSHALL , BHOGAL KULVIR , GALE MARTIN , HAMILTON RICK ALLEN II
IPC: H04L12/58
Abstract: Embodiments of the present invention provide an approach for an enhanced pub/sub environment whereby a peer group of publishers is defined and can collaborate to determine when to cache data being sent to a common broker or brokers. Specifically, in a typical embodiment, configuration data or the like will be read to define a peer group among the publisher systems in the pub/sub messaging environment. Once such a group has been identified, communication therebetween will be established. Once cacheable data has been identified, a unique identifier/set of cache keys will be generated corresponding thereto and associated with the set of data. A message comprising the set of data and corresponding unique identifier will then be sent to a broker system (e.g., from the publisher system(s)), which will store (or have stored) the set of data in a cache, and pass the underlying message along to applicable subscriber system(s).
-
公开(公告)号：GB2501656A
公开(公告)日：2013-10-30
申请号：GB201314915
申请日：2012-02-22
Applicant: IBM
Inventor： JONES ANDREW R , O'CONNELL BRIAN MARSHALL , BHOGAL KULVIR , GALE MARTIN , HAMILTON RICK ALLEN II
IPC: H04L12/58
Abstract: Embodiments of the present invention provide an approach for an enhanced pub/sub environment whereby a peer group of publishers is defined and can collaborate to determine when to cache data being sent to a common broker or brokers. Specifically, in a typical embodiment, configuration data or the like will be read to define a peer group among the publisher systems in the pub/sub messaging environment. Once such a group has been identified, communication therebetween will be established. This communication will be utilized to identify a set of data that is common to the set (at least one) of publisher systems in the peer group that has a potential to be cacheable. Once such data has been identified, a unique identifier/set of cache keys will be generated corresponding thereto and associated with the set of data. A message comprising the set of data and corresponding unique identifier will then be sent to a broker system (e.g., from the publisher system(s)), which will store (or have stored) the set of data in a cache, and pass the underlying message along to applicable subscriber system(s). When a future message is to be generated with the set of data previously cached, the publisher system will instead substitute the unique identifier for the cached set of data to yield an abbreviated message, which is sent to the broker system. Upon receipt of the abbreviated message, the broker system will locate the unique identifier, retrieve the corresponding set of data from cache, substitute the set of data for the unique identifier to yield a completed message, and send the completed message to the applicable subscriber system(s).
-
Patent Agency Ranking
公开(公告)号：DE112011103081T5
公开(公告)日：2013-09-12
申请号：DE112011103081
申请日：2011-08-31
Applicant: IBM
Inventor： AMSTERDAM JEFFREY DAVID , JENSEN BRIAN WILLIS , WALKER KEITH RAYMOND , O'CONNELL BRIAN MARSHALL
Abstract: Ein computerrealisiertes Verfahren, System und Computerprogrammprodukt für das Verwalten von Clientverbindungen aus einer Vielzahl von Clients mit einem Server. Ein Nachrichtenübertragungsserver erkennt durch einen ersten Server ein Attribut einer Clientverbindung, das einen Schwellenwert übersteigt, wobei das Attribut eine Schreibauswahl-Operation ist. Als Reaktion auf das Erkennen, dass das Attribut des Clients den Schwellenwert von Schreibauswahl-Aufrufen übersteigt, wählt der Nachrichtenübertragungsserver einen Client aus, der eine Verbindung mit einem Server herstellt. Als Reaktion auf das Auswählen des Clients wählt der Nachrichtenübertragungsserver einen Verbindungsmechanismus eines zweiten Servers aus. Der Nachrichtenübertragungsserver überträgt eine Verlagerungsnachricht an den Client, wobei die Verlagerungsnachricht mindestens einen einheitlichen Ressourcenbezeichner (Uniform Resource Identifier, URI) aufweist.
-
公开(公告)号：DE112012000989B8
公开(公告)日：2018-07-26
申请号：DE112012000989
申请日：2012-02-22
Applicant: IBM
Inventor： JONES ANDREW , O'CONNELL BRIAN MARSHALL , BHOGAL KULVIR SINGH , GALE MARTIN , HAMILTON II RICK
-
公开(公告)号：CA2673950C
公开(公告)日：2015-03-31
申请号：CA2673950
申请日：2008-04-11
Applicant: IBM
Abstract: Generally speaking, systems, methods and media for authenticating a user to a server based on previous authentications to other serversare disclosed. Embodiments of amethod for authenticating a user to a servermay include receiving a request to authenticate the user to the server and determining whether authenticating the user requires matching an authentication plan.If a plan is required, the method may also include accessing a stored authentication plan with authentication records each having expected information relating to user access to a different server. The method may also include receiving an indication of the user s current authentication plan from an authentication store where the plan has authorization records each having current information relating to user access. Embodiments ofthe method may also include comparing the stored authentication plan with the received current authentication plan to determine whether they match and, in response to a match, authenticating the user.
-
公开(公告)号：DE112011103498T5
公开(公告)日：2013-07-25
申请号：DE112011103498
申请日：2011-12-13
Applicant: IBM
Inventor： O'CONNELL BRIAN MARSHALL , WALKER KEITH RAYMOND
IPC: H04L47/762
Abstract: Ausführungsformen zur Optimierung einer Server-Verbindung in einer dauerhaft verbundenen Veröffentlichen/Abonnieren-Umgebung werden bereitgestellt. Ein Computer ordnet eine erste Aktivitätskategorie einem Client zu. Der Computer protokolliert eine oder mehrere Auswahloperationen einer Verbindung zwischen einem Client und einem ersten Server in einer Protokolldatei. Der Computer berechnet ein Verhältnis einer Anzahl der Auswahloperationen zu einer Überwachungszeit des ersten Servers, wobei die Überwachungszeit ein Zeitintervall ist, das seit einer vorangegangenen Überprüfung der protokollierten Daten vergangen ist. Der Computer legt ausgehend von einem Ablesen des Verhältnisses aus einer Verhältnisbereichsliste in einer Steuerdatei eine zweite Aktivitätskategorie für den Client fest, wobei die zweite Aktivitätskategorie nicht die erste Aktivitätskategorie ist. Als Reaktion auf das Festlegen der zweiten Aktivitätskategorie verschiebt der Computer den Client von dem ersten Server in einen zweiten Server.
-
公开(公告)号：CA2644956A1
公开(公告)日：2007-10-18
申请号：CA2644956
申请日：2007-03-01
Applicant: IBM
Inventor： WALKER KEITH RAYMOND , O'CONNELL BRIAN MARSHALL
IPC: G06F21/20
-
-
-
-
-
-
-
-
Search Results
19
records
(took 0.018 seconds)
