## Lecture-8

• If an algorithm can perform fruitless actions indefinitely with finite probability, the algorithm is said to be unstable. • Activity is initiated by an overloaded node (sender)• A task is sent to an underloaded node (receiver) • A node is identified as a sender if a new task originating at the node makes the queue length exceed a threshold T.
• Only new arrived tasks are considered for transfer • Random: dynamic location policy, no prior information exchange• Threshold: polling a node (selected at random) to find a receiver• Shortest: a group of nodes are polled to determine their queue • Location policies adopted cause system instability at high loads Yes QueueLength at "I"
Arrives

• Initiated from an underloaded node (receiver) to obtain a task from an overloaded node (sender) • A node selected at random is polled to determine if transferring a task from it would place its queue length below the threshold level, if not, the polled node transfers a task.
• Do not cause system instability in high system load, however, in • Most transfers are preemptive and therefore expensive Yes QueueLength at "I"
• Thresholds are equidistant from the node's estimate of the • Sender-initiated component: Timeout messages TooHigh, TooLow, Accept, AwaitingTask, ChangeAverage • Receiver-initiated component: Timeout messages TooLow, LooHigh, Accept, AwaitingTask, ChangeAverage • Similar to both the earlier algorithms • A demand-driven type but the acceptable range can be increased/decreased by each node individually.
• A Stable Symmetrically Initiated Algorithm - Utilizes the information gathered during polling to classify the nodes in the system as either Sender, Receiver or OK.
- The knowledge concerning the state of nodes is maintained by a data structure at each node, comprised of a senders list, a receivers list, and an OK list.
- Initially, each node assumes that every other node is a receiver.
- Transfer Policy • Triggers when a new task originates or when a task departs.
• Receiver-initiated transfers require - Task Placement refers to the transfer of a task that is yet to begin execution to a new location and start its execution there.
- Task Migration refers to that transfer of a task that has already begun execution to a new location and continuing its execution there.
• The transfer of the task's state including information e.g. registers, stack, ready/blocked, virtual memory address space, file descriptors, buffered messages etc. to the new machine.
• The task is frozen at some point during the transfer so that the state does not change further.
• The task is installed at the new machine and is put in the ready queue so that it can continue executing.
- To support remote execution, obtaining and transferring the state, and Residual Dependencies
- Refers to the amount of resources a former host of a preempted or migrated task continues to dedicate to service requests from the migrated task.
Implementations
• Attempts to reduce the freezing time of a migrating task by precopying the state.
• The bulk of the task state is copied to the new host• It increases the number of messages that are sent to new host • Makes use of the location-transparent file access mechanism provided by its file • All the modified pages of the migrating task are swapped to file server • Reduction in migration is achieved by using a feature called Copy-on-Reference• The entire virtual memory address space is not copied to the new host • Services that are provided to user processes irrespective of the location of the processes and services.
• In distributed systems, it is essential that the location • Location transparency in principle requires that names (e.g. process names, file names) be independent of their location (i.e. host names).
• Any operation (such as signaling) or communication that was possible before the migration of a task should be possible after its migration • Example - SPRITE - Location Transparency Mechanisms - A location-transparent distributed file system is provided- The entire state of the migrating task is made available at the new host, and therefore, any kernel calls made will be local at new host - Location-dependent information such as host of a task is maintained • Issues involved in Migration Mechanisms - Decision whether to separate the policy-making modules • It has implications for both performance and the ease of • The separation of policy and mechanism modules simplifies the - Decision to where the policy and mechanisms should • The migration mechanism may best fit inside the kernel• Policy modules decide whether a task transfer should occur, this - Interplay between the task migration mechanism and • The mechanisms can be designed to be independent of one another so that if one mechanism's protocol changes, the other'sneed not • Comparing the performance of task migration mechanisms implemented in different systems is a difficult task, because of the different, • SPRITE consists of a collection of SPARCSTATION 1• CHARLOTTE consists of VAX/11-750 machines - Operating systems- IPC mechanism- File systems- Policy mechanisms

Source: http://www.ssuet.edu.pk/courses/ce403/AllSec/Lectures/Lecture-8.pdf

### Hardy

New developments in aquatic feed ingredients, and potential of Hagerman Fish Culture Experiment Station, University of Idaho, 3059F National Fish Hatchery Road, Hagerman, ID 83332, USA ABSTRACT: Aquaculture production has expanded at a rate of 15% per year and is predicted to continue to grow at this rate for at least the next decade. Demands on traditional fish feed ingredients, mainly fish

### Audretsch - universitäten und regionales wirtschaftswachstum

Max-Planck-Institut Ökonomik Universitäten und regionales Wirtschaftswachstum Group Entrepreneurship, Growth and Public Policy Universitäten und regionales Wirtschaftswachstum Eine stille und praktisch unbemerkte Veränderung hat die Wirtschaftspolitik ergriffen. Wäh-rend sich die Sicherung von Wirtschaftswachstum und die Schaffung von Arbeitsplätzen bis-her auf fiskalische