Path-based scheduling

Canq>osano, R. (1991) Path-Based Scheduling for Synthesis. IEEE Transactions on Computer-Aided Design, Vol 10, No 1, pp 85-93. 

R. Camposano. Path-based scheduling for synthesis. IEEE Transaction on CAD, CAD-10, pages 85-93, Jan 1991. 

Amical uses a dynamic loop scheduling algorithm [13, 12]. This algorithm is adapted to control-flow-dominated designs written in Vhdl, and is a development of the path-based approach proposed by Camposano [2]. Essentially, the scheduler reads in a Vhdl description and produces a behavioral FSM in the form of a transition table. Each transition (macro-cycle) corresponds to the execution of a control step under a given condition. A macro-cycle may need several basic cycles (clock cycles) for its execution. The top left window in figure 4 shows the transition table composed of two states and five transitions. 

R. Bergamaschi, R. Camposano, and M. Payer, "Area and Performance Optimizations in Path-Based Scheduling, Proc. ofEDAC 91, Amsterdam, February 1991. 

R. Camposano, Path-Based Scheduling for Synthesis, Proceedings of the HICSS 90, pp. 348-355, Kailua-Kona, Hawaii, January 1990. 

Heuristic path-based scheduling First, all possible execution paths through the controlflow graph are found, and timing and... 

Uses AFAP path-based scheduling, as described above. After the paths are merged, where there is more than one possible cut, the dataflow analysis is used to choose the cut with the smallest number of live values, thus minimizing the number of registers. 

Raul Camposano, "Path-Based Scheduling for Synthesis Trans, on CAD, pages 85-93, January 1991. 

AFAP path-based scheduling, the Intel 8251, a Kalman filter example, a greatest common divisor algorithm example, a counter example, and others. Essentially an updated version of Camposano90a. 

Raul Camposano and Reinaldo A. Bergamaschi, S3mthesis Using Path-Based Scheduling Algorithms and Exercises , Proc. of the 27th DAC, pages 450-455, June 1990. 

Heuristic path-based scheduling First, all possible execution paths through the controlflow graph are found, and timing and resource constraints are expressed as intervals where the paths must be cut to separate the operations into states. Then heuristics are used to compute a cutting point, and cut all paths containing that point. 

HIS uses some ideas pioneered in the YSC, specially in the area of path-based scheduling [12]. Most algorithms however are new, improving the quality of the resulting designs. Moreover, HIS puts much more emphasis on practicality. VHDL [26], by now an established standard, was chosen as the input language... 

Different solutions to this problem have been proposed. In [7], Fisher described a series of transformations used in trace scheduling for microcode compaction. This method compacts the critical path independently of the control block boundaries. A refinement of this tqjproach was proposed recently by Potasman et al and is called percolation scheduling [8]. Camposano and Ber unaschi use an alternative approach called path-based scheduling [9], wh e all possible paths are compacted independently. A post-processing step merges separate paths to Iowa the controller costs. 

Most detailed programs for plant-related projects are prepared using simple critical path analysis (CPA) techniques. CPA is a well-known and familiar technique to most engineers and is incorporated into most modern PC-based planning and scheduling systems. Such a system typically has the following features that a plant project manager will find useful ... 

The process of updating the schedule is basically the same as that used in planning the shutdown. It is simply based on new information and deals only with the remaining portion of the shutdown. In some very large shutdowns, it may be desirable to perform time-cost trade-off analysis, resource leveling, etc., each time the shutdown CPM is updated. In many cases, however, the update will involve only the basic forward, backward and slack computations. These calculations will allow for development of a revised schedule. They will also identify the critical path that is likely to change over the course of the shutdown. 

The schedules shown to the committee were simply lists of things that had to be done and the time estimated to complete each of them. Alternative paths that would be generated as decisions were made were not evident, nor were any time-related probabilities assigned. The schedules presented to the committee are extremely vulnerable to unforeseen or external events. Moreover, the committee was not provided with total project cost estimates, even of a conceptual or preliminary nature. Without these estimates, it is extremely difficult to evaluate engineering decisions and select the most cost-effective technologies. Based on the lack of documentation, the committee concluded that an adequate preproject plan for a modified baseline process at Pueblo may not exist. If this is true, the completion of the project on schedule and within budget is unlikely. 

The proposed schedule for commercialization targets 1985 for start of production. This assumes that all environmental and permitting requirements will be met in time for construction to start in mid-1982. Based on past experience, this is an adequate time for preparation of an EIS and for PSD review but these activities are on the critical path so any additional time spent on unusual problems would certainly result in a day-for-day slippage in mechanical completion. 

Network logic determines the relationship between the tasks or team member deliverables, either sequential (finish to start) or overlapping (start to start and/or finish to finish). There are numerous software packages available at low cost to support calculation of the critical path and float. However, the real challenge of project management tools is the creation of the project network based upon historical experience and team involvement. By definition projects are rmique and, in the pharmaceutical arena, each project is different. Therefore, the importance of team member involvement in the development, tracking, and adjustment of the project schedule is crucial to successful application of project management basics. 

The detail schedule should provide a calendar-based schedule of activities, tasks, and key events of the master schedule. The detail schedule should be used to track the progress of the technical effort. Detail schedule data should be used to construct a network of events, tasks, and activities to determine the critical path of the engineering efforts and to analyze variances to schedule. 

We have presented a refinement-based proof-method for the verification of modern processor architectures, and demonstrated its applicability by showing the correctness of a data-path involving multiple functional units, registerrenaming, dynamic scheduling, and out-of-order execution, thus significantly exceeding the complexitiy of static pipelines as e.g. treated in [BD94]. Still, this work needs to be extended in a number of ways. 

---