A Discriminatory Pay-as-Bid Mechanism for Efficient Scheduling in the Sun N1 Grid Engine

Low-energy automated scheduling of computing resources

“…The allocation is primarily handled by a sealed bid auction mechanism, where requesters and suppliers do not know the other users' requests and offers. (See also [32] and [52].…”

“…Consolidated attempts are present in the work of [23][24] and [33] ranging from automated power management, to forcing servers to 'power-nap' in underutilized phases.When it comes to pricing goods with incomplete information, the case generally incurred when allocating computing resources, auctions have often been a part of the solution in some way or another. A foundation was laid by [3] based on which several authors have covered in detail Grid market models ranging from commodity exchanges [58] through combinatorial double auctions [46] to heuristics [28] and [52]. Mostly, CPU power from the nodes have been auctioned to job requesters, with the agents requesting jobs paying the node suppliers through a Scheduling agent which allocates the job to a specific node.…”

“…Heuristics generally form a distinct converse trade-off between computational speed and allocative efficiency. In order to determine the behavior of such heuristic solutions, we need to rigorously analyze both the optimal allocation mechanism, and the heuristic to come to valuable insights [52].…”

Delivering IT as A Utility- A Systematic Review

“…But in recent years, Grid computing has emerged as a fast developing technology with the capability to surmount the bottleneck of inability to utilize the capability of physically dispersed resources. Grid computing is a promising concept to enhance the effectiveness of existing computing systems and to cut down on IT expenditures by providing dynamic access to computer resources across geographical and organizational boundaries [47]. Grid is a type of parallel and distributed system that enables the sharing, selection, and aggregation of geographically distributed autonomous resources dynamically at runtime depending on their availability, capability, performance, cost, and users QoS requirements [48] [49].…”

SORMA – Business Cases for an Open Grid Market: Concept and Implementation

“…Prominent examples of market mechanisms for scheduling of computational resources like CPU and Memory are based on proportional share mechanisms [15], where the users receive a share of the computer resource proportional to their valuations fraction of the overall valuation across all users. A related and implemented mechanism is the so-called pay-as-bid mechanism proposed in [23,26], where the user pays the price he has bid.…”