-
公开(公告)号：DE60024611T2
公开(公告)日：2006-07-27
申请号：DE60024611
申请日：2000-05-09
Applicant: IBM
Inventor： DROZ PATRICK , VAN LUNTEREN JAN
IPC: G06F12/02 , H04L12/701 , H04L12/743 , H04L12/747
Abstract: A system is provided for finding longest matching prefixes of search arguments such as IP addresses, in a multiple-level structure of node tables. Segments of a given search argument each having individually determined length are selected, and some segments are used as test argument to be compared to a stored test value, using the test result for determining a further lookup step, whereas other segments are used as index value for accessing an entry in a node table; also, a combination of testing followed by indexed accessing with such individually selected search argument segments can be used during the lookup procedure. Thus, it is possible to optimally adapt the used node table data structure and the search procedure for longest matching prefixes, to the actual distribution of prefixes over the whole address space. Two-stage addressing of entries is also provided, i.e. a node table is selected by a table base pointer, and the entry selection within the node table will be effected in response to a test result or an index operation, determining the offset within the table.
-
公开(公告)号：DE60024611D1
公开(公告)日：2006-01-12
申请号：DE60024611
申请日：2000-05-09
Applicant: IBM
Inventor： DROZ PATRICK , VAN LUNTEREN JAN
IPC: G06F12/02 , H04L12/701 , H04L12/743 , H04L12/747
Abstract: A system is provided for finding longest matching prefixes of search arguments such as IP addresses, in a multiple-level structure of node tables. Segments of a given search argument each having individually determined length are selected, and some segments are used as test argument to be compared to a stored test value, using the test result for determining a further lookup step, whereas other segments are used as index value for accessing an entry in a node table; also, a combination of testing followed by indexed accessing with such individually selected search argument segments can be used during the lookup procedure. Thus, it is possible to optimally adapt the used node table data structure and the search procedure for longest matching prefixes, to the actual distribution of prefixes over the whole address space. Two-stage addressing of entries is also provided, i.e. a node table is selected by a table base pointer, and the entry selection within the node table will be effected in response to a test result or an index operation, determining the offset within the table.
-
公开(公告)号：BR0317582A
公开(公告)日：2005-11-22
申请号：BR0317582
申请日：2003-10-30
Applicant: IBM
Inventor： BOWEN ED , DROZ PATRICK , JEFFRIES CLARK D , KENCL LUKAS , KIND ANDREAS , PLETKA ROMAN A , MANNAL SOENKE
IPC: H04L12/56
Abstract: Methods and apparatus are provided for controlling flow rates of a plurality of data packet flows into a queue 4 corresponding to a resource 3 of a network device 1 . The flows comprise a set 7 of non-responsive flows, and a set 8 of other flows which may comprise responsive flows and/or flows whose responsiveness is unknown. The flow rates are managed in accordance with a queue management scheme such that adjustments are made to each flow rate in dependence on excess bandwidth in the resource, the amounts of the adjustments being dependent on one or more adjustment parameters for each flow. An error signal is generated based on the deviation from a desired allocation ratio of the ratio of the total flow rates into the queue 4 for the sets of flows 7, 8 . At least one adjustment parameter for at least one flow is then varied in dependence on the error signal in such a manner as to reduce the aforementioned deviation. A closed-loop control scheme thus operates in conjunction with the underlying queue management scheme to promote fair bandwidth allocation even in the presence of a mix of responsive and non-responsive flows.
-
公开(公告)号：CA2509409C
公开(公告)日：2009-04-21
申请号：CA2509409
申请日：2003-10-30
Applicant: IBM
Inventor： MANNAL SOENKE , BOWEN ED , KIND ANDREAS , PLETKA ROMAN A , JEFFRIES CLARK D , DROZ PATRICK , KENCL LUKAS
IPC: H04L12/56
Abstract: Methods and apparatus are provided for controlling flow rates of a plurality of data packet flows into a queue 4 corresponding to a resource 3 of a netwo rk device 1. The flows comprise a set 7 of non-responsive flows, and a set 8 of other flows which may comprise responsive flows and/or flows whose responsiveness is unknown. The flow rates are managed in accordance with a queue management scheme such that adjustments are made to each flow rate in dependence on excess bandwidth in the resource, the amounts of the adjustmen ts being dependent on one or more adjustment parameters for each flow. An error signal is generated based on the deviation from a desired allocation ratio o f the ratio of the total flow rates into the queue 4 for the sets of flows 7, 8. At least one adjustment parameter for at least one flow is then varied in dependence on the error signal in such a manner as to reduce the aforementioned deviation. A closed-loop control scheme thus operates in conjunction with the underlying queue management scheme to promote fair bandwidth allocation even in the presence of a mix of responsive and non- responsive flows.
-
公开(公告)号：DE60313037T2
公开(公告)日：2007-12-13
申请号：DE60313037
申请日：2003-10-30
Applicant: IBM
Inventor： BOWEN ED , DROZ PATRICK , JEFFRIES CLARK D , KENCL LUKAS , KIND ANDREAS , PLETKA ROMAN A , MANNAL SOENKE
IPC: H04L12/56
Abstract: Methods and apparatus are provided for controlling flow rates of a plurality of data packet flows into a queue 4 corresponding to a resource 3 of a network device 1 . The flows comprise a set 7 of non-responsive flows, and a set 8 of other flows which may comprise responsive flows and/or flows whose responsiveness is unknown. The flow rates are managed in accordance with a queue management scheme such that adjustments are made to each flow rate in dependence on excess bandwidth in the resource, the amounts of the adjustments being dependent on one or more adjustment parameters for each flow. An error signal is generated based on the deviation from a desired allocation ratio of the ratio of the total flow rates into the queue 4 for the sets of flows 7, 8 . At least one adjustment parameter for at least one flow is then varied in dependence on the error signal in such a manner as to reduce the aforementioned deviation. A closed-loop control scheme thus operates in conjunction with the underlying queue management scheme to promote fair bandwidth allocation even in the presence of a mix of responsive and non-responsive flows.
-
Patent Agency Ranking
公开(公告)号：MY133313A
公开(公告)日：2007-11-30
申请号：MYPI20034435
申请日：2003-11-19
Applicant: IBM
Inventor： BOWEN HOYT EDWIN JR , DROZ PATRICK , JEFFERIES CLARK D , KENCL LUKAS , KIND ANDREAS , MANNAL SOENKE V , PLETKA ROMAN A
IPC: H04L12/56
Abstract: METHODS AND APPARATUS ARE PROVIDED FOR CONTROLLING FLOW RATES OF A PLURALITY OF DATA PACKET FLOWS INTO A QUEUE 4 CORRESPONDING TO A RESOURCE 3 OF A NETWORK DEVICE 1. THE FLOWS COMPRISE A SET 7 OF NON-RESPONSIVE FLOWS, AND A SET 8 OF OTHER FLOWS WHICH MAY COMPRISE RESPONSIVE FLOWS AND/OR FLOWS WHOSE RESPONSIVENESS IS UNKNOWN. THE FLOW RATES ARE MANAGED IN ACCORDANCE WITH A QUEUE MANAGEMENT SCHEMESUCH ADJUSTMENTS ARE MADE TO EACH FLOW RATE IN DEPENDENCE ON EXCESS BANDWIDTH IN THE RESOURCE, THE AMOUNTS OF THE ADJUSTMENTS BEING DEPENDENT ON ONE OR MORE ADJUSTMENT PARAMETERS FOR EACH FLOW. AN ERROR SIGNAL IS GENERATED BASED ON THE DEVIATION FROM A DESIRED ALLOCATION RATIO OF THE RATIO OF THE TOTAL FLOW RATES INTO THE EQUEUE 4 FOR THE SETS OF FLOWS 7, 8. AT LEAST ONE ADJUSTMENT PARAMETER FOR AT LEAST ONE FLOW IS THEN VERIED IN DEPENDENCE ON THE ERROR SIGNAL IN SUCH A MANNER AS TO REDUCE THE AFOREMENTIONED DEVIATION. A CLOSED-LOOP CONTROL SCHEME THUS OPERATES IN CONJUCTION WITH THE UNDERLYING QUEUE MANAGEMENT SCHEME TO PROMOTE FAIR BANDWIDTH ALLOCATION EVEN IN THE PRESENCE OF A MIX OF RESPONSIVE AND NON-RESPONSIVE FLOWS.(FIG. 1)
-
公开(公告)号：DE60303384D1
公开(公告)日：2006-04-13
申请号：DE60303384
申请日：2003-06-20
Applicant: IBM
Inventor： BOWEN EDWIN , DROZ PATRICK , KENCL LUKAS , METZLER BERNHARD
Abstract: Load balancing apparatus for a data communications network comprises hash logic for computing a hash function on incoming data packets. A threshold detector is connected to the hash logic for triggering, in response to utilization of the downstream objects exceeding a predefined threshold, redefinition in the hash logic of parameters of the hash function from a first set of parameters to a second set of parameters for redistributing the data packets amongst the downstream objects. In use, the hash logic, directs the packets for routing to downstream objects in the network via a first routing path based on a hash computation using the first set of parameters, and, if the threshold is exceeded, selectively directs the packets to one of the first routing path and a second routing path in dependence on separate hash computations using the first and the second sets of parameters for subsequent routing of the packets via the selected one of the first and second routing paths based on the results of one of the separate hash computations.
-
公开(公告)号：DE69735821D1
公开(公告)日：2006-06-08
申请号：DE69735821
申请日：1997-06-25
Applicant: IBM
Inventor： DROZ PATRICK , DYKEMAN DOUGLAS
-
公开(公告)号：DE69533760T2
公开(公告)日：2005-11-03
申请号：DE69533760
申请日：1995-12-13
Applicant: IBM
Inventor： DROZ PATRICK
Abstract: In a high-speed packet switched communication system such as an ATM network, the number of existing connections is limited by a Connection Admission Control (CAC) procedure. For each newly requested connection, the expected traffic rate is added to an already reserved bandwidth and a test is made whether the sum is above a given limit.According to the invention, the reserved bandwidth (r_bw) is updated not only for each new connection adding the additional requested capacity (RQ), but also in-between using an effective traffic capacity value derived from measuring the actual traffic (EC). For obtaining a reasonable effective traffic capacity value, the traffic sample sequence is modified by filtering out the "noise" representing short-time variations. This is done in an iterative procedure in which the cutoff between the signal and the "noise" portion is adaptively and dynamically changed so that the available buffer capacity can accommodate just that noise portion. For the filtering process, wavelet transformation is used.
-
公开(公告)号：DE69533760D1
公开(公告)日：2004-12-16
申请号：DE69533760
申请日：1995-12-13
Applicant: IBM
Inventor： DROZ PATRICK
Abstract: In a high-speed packet switched communication system such as an ATM network, the number of existing connections is limited by a Connection Admission Control (CAC) procedure. For each newly requested connection, the expected traffic rate is added to an already reserved bandwidth and a test is made whether the sum is above a given limit.According to the invention, the reserved bandwidth (r_bw) is updated not only for each new connection adding the additional requested capacity (RQ), but also in-between using an effective traffic capacity value derived from measuring the actual traffic (EC). For obtaining a reasonable effective traffic capacity value, the traffic sample sequence is modified by filtering out the "noise" representing short-time variations. This is done in an iterative procedure in which the cutoff between the signal and the "noise" portion is adaptively and dynamically changed so that the available buffer capacity can accommodate just that noise portion. For the filtering process, wavelet transformation is used.
-
-
-
-
-
-
-
-
-
Search Results
22
records
(took 0.018 seconds)
