International Business Machines Corporation

Cisco Technology, Inc.

Jean A Lorrain

Kiyoshi Maruyama

German Sergio Goldszmidt

Pascal Thubert

Dinesh C Verma

Didier F Giroir

Eric Michel Levy-Abegnoli

Richard Harold Boivie

Jarir Kamel Chaar

Daniel Manuel Dias

Colin George Harrison

David Alson George

Jean-Pierre Marce

Shun Shing Chan

Miram Zohar

Karen Appleby-Hougham

Pascal Pol Marie Thubert

A method (and system) for managing and controlling allocation and de-allocation of resources based on a guaranteed amount of resource and additional resources based on a best effort for a plurality of customers, includes dynamically allocating server resources for a plurality of customers, such that the resources received by a customer are dynamically controlled and the customer receives a guaranteed minimum amount of resources as specified under a service level agreement (SLA). The best effort is defined in the SLA as a range of service to be provided to the customer if server resources are currently available.

Method and apparatus for dynamically adjusting resources assigned to plurality of customers, for meeting service level agreements (SLAs) with minimal resources, and allowing common pools of resources to be used across plural customers on a demand basis

Today, a Client and an associated Server use a single IP priority for all the data packets exchanged over a TCP session used to transport the Telnet traffic (file transfer and screen related interactions). The present invention discloses a method and system to provide the Client with the capability, when exchanging a file with an SNA Host over a TCP session, to dynamically change (i.e downgrade) the IP priority associated to all data packets flowing to the TN3270 Server over the TCP session, during the whole duration of the file transfer. The associated Server, on receipt of a data packet from the client (due to screen related interactions or to file transfers), memorizes the associated priority and uses this priority over all packets that it sends back to the client. This allows IP network implementation to give higher priority treatment to the data packets due to screen related interactions (interactive traffic) over data packets related to file transfer (batch traffic). The priority used by data packets related to file transfer is customisable, to give maximum freedom to the network designer. For example, SNA true interactive traffic (due to screen related interactions) could flow on the highest priority, pure IP traffic could flow on any of the possible medium priorities and data packet due to file transfer could use the lowest priority within the network.

System using IP transmission priority for improving overall network response time in TELNET 3270 sessions

A method (and system) for managing and controlling allocation and de-allocation of resources based on a guaranteed amount of resource and additional resources based on a best effort for a plurality of customers, includes dynamically allocating server resources for a plurality of customers, such that the resources received by a customer are dynamically controlled and the customer receives a guaranteed minimum amount of resources as specified under a service level agreement (SLA).

Method in a telnet client for transmitting in a session, file traffic to an application through a telnet server in an Internet protocol (IP) network. The method includes assigning by the telnet client an IP transmission priority to file traffic different from the IP transmission priority used for interactive traffic and transmitting the file traffic to the telnet server using said assigned IP transmission priority.

System using IP transmission priority for improving overall network response time in telnet 3270 sessions

METHOD AND APPARATUS FOR DYNAMICALLY ADJUSTING RESOURCES ASSIGNED TO PLURALITY OF CUSTOMERS, FOR MEETING SERVICE LEVEL AGREEMENTS (SLAS) WITH MINIMAL RESOURCES, AND ALLOWING COMMON POOLS OF RESOURCES TO BE USED ACROSS PLURAL CUSTOMERS ON A DEMAND BASIS

A system and method for controlling the rates at which application workload, e.g., TCP connection requests, is admitted to a collection of servers, such as a server farm of an application service provider (ASP) that hosts Internet World Wide Web (WWW) sites of various owners. The system and method are intended to operate in an environment in which each customer has a workload-based SLA for each type of application hosted by the provider and used by the customer. The system and method achieve support (minimum, maximum) TCP connection requests for multiple customers and applications. According to one aspect, the system and method guarantee, control and deliver TCP connection-based workload SLA's to customers whose applications are hosted by the server farm with the use of a workload regulator that operates by regulating only new TCP connection request packets while transparently passing existing TCP connection packets and other request packets received for customers. The regulator further operates by regulating the flow of incoming TCP connection requests to each customer business activity application so as to guarantee a level of service previously agreed to each customer (per their respective SLA's) by applying rate admittance to requests and by dropping (or rejecting) requests to guarantee the agreed service levels to the customer's application.

System and method for regulating incoming traffic to a server farm

A highly scalable system and method for supporting (mim,max) based Service Level Agreements (SLA) on outbound bandwidth usage for a plurality of customers whose applications (e.g.,Web sites) are hosted by a server farm that consists of a very large number of servers. The system employs a feedback system that enforces the outbound link bandwidth SLAs by regulating the inbound traffic to a server or server farm. Inbound traffic is admitted to servers using a rate denoted as Rt(i,j), which is the amount of the icustomer's jtype of traffic that can be admitted within a service cycle time to servers which support the icustomer. A centralized device computes Rt(i,j) based on the history of admitted inbound traffic to servers, the history of generated outbound traffic from servers, and the SLAs of various customers. The Rt(i,j) value is then relayed to one or more inbound traffic limiters that regulate the inbound traffic using the rates Rt(i,j) in a given service cycle time. The process of computing and deploying Rt(i,j) values is repeated periodically. In this manner, the system provides a method by which differentiated services can be provided to various types of traffic, the generation of output from a server or a server farm is avoided if that output cannot be delivered to end users, and revenue can be maximized when allocating bandwidth beyond the minimums.

Highly scalable system and method of regulating internet traffic to server farm to support (min,max) bandwidth usage-based service level agreements

A method and system for controlling and guaranteeing a service level agreement (SLA) based on a communications outbound link bandwidth usage to a plurality of customers having electronic business activity hosted by at least one server as a server farm, includes monitoring the outbound communications bandwidth usage by each customer traffic to determine a level of service being provided to each customer with respect to the agreed service level agreement in each service cycle time per unit of time. The flow of incoming requests to each customer business activity application is controlled so as to guarantee a level of service previously agreed to the customer by queuing requests to the customer and by selectively dropping requests to the customer to guarantee the agreed service levels to the customer. The controlling process controls and guarantees each outbound link usage based service level agreement by controlling the flow of incoming requests to the at least one server.

System and method for enforcing communications bandwidth based service level agreements to plurality of customers hosted on a clustered web server

A system, method and article of manufacturing for increasing link bandwidth occupation in a high speed packet switching digital network by enabling merging the traffics provided by different source users over several network node entry ports and to be propagated throughout network paths toward a same destination network port. To that end, at network ingress, the original packets provided by said source users and entering the network, are encapsulated with a so-called Single Sided Virtual Channel (SSVC) header including a Data Link Connection Identification (DLCI) field. Then, the packets provided by said source users and entering a given network node along their predefined path are monitored. Said packets SSVC headers DLCI fields are loaded with a same Virtual Channel number, whereby the corresponding traffics are being merged into a same channel, down to the destination network node. Then, the packets in said destination node are de-encapsulated from said SSVC header.

System, method, and article of manufacture for increasing link bandwidth utilization in a high speed digital network

A method for creating a single virtual LAN including a source (S) attached to a first Token ring LAN (N ), a target (T) within a subnet attached to a different Token ring LAN (N ), and an interconnecting router (R). The source (S) encapsulates the first packet with a Token ring header including the router MAC address (RMAC), SMAC as source MAC address as layer 2 information, and IP address of T (TIP) as layer data and sends this packet over N toward the router. When receiving the first packet, R reads its IP table for best match with TIP address to identify the subnet including T. The net handler runs an ARP protocol to identify TMAC address, substitutes the MAC header with said TMAC address into said first packet destination MAC address field and forwards said first packet over N . R then sends a conventional ICMP message over N limited broadcast and reconfigures itself in proxy ARP for the defined subnet. For the second packet, S runs an ARP and R answers with RMAC address and the RIF data to be used for next packets.

Method and system for improving high speed internetwork data transfers

A method and system for improving traffic operation in an internet environment by speeding up data packet transfers between a source host attached to a first Local Area Network (LAN), and a target host attached to a second (different) LAN, both LANs being interconnected by a router. The invention enables said router self configuring into a bridge over the path between said source and target hosts, selectively and dynamically during traffic operation. To that end, upon first packet being sent, the source host pushes said packet over the first LAN toward the router. When receiving said first packet, the router reads its IP table locating the target host on its LAN, and then runs a conventional ARP over its LAN to get the address of the target sending back its MAC address. The router stores the MAC address of the target host into an ARP table and sets an entry into a so-called transparent bridging table accordingly. The router reconfigures itself into Proxy-ARP to simulate it is the target host, and sends the first packet over the second toward the target host. The router applies then a conventional ICMP over the first LAN to get all hosts (including the source host) located on the first LAN updating their IP routing table to simulate the target host as being attached to the first LAN. Before sending next packet, the source host runs an ARP protocol over the first LAN. The router answers with MAC address for the target host. The source sends this second packet to IP address for the target host with MAC address for the target host over the first LAN, and the router bridges.

Growing community of inventors

Vence, France

Jean A Lorrain