Heeheon Kim

Chameleon: A High Performance Flash/FRAM Hybrid Solid State Disk Architecture

Flash memory solid state disk (SSD) is gaining popularity and replacing hard disk drive (HDD) in mobile computing systems such as ultra mobile PCs (UMPCs) and notebook PCs because of lower power consumption, faster random access, and higher shock resistance. One of the key challenges in designing a high-performance flash memory SSD is an efficient handling of small random writes to non-volatile data whose performance suffers from the inherent limitation of flash memory that prohibits in-placc update. In this paper, we propose a high performance Flash/FRAM hybrid SSD architecture called Chameleon. In Chameleon, metadata used by the flash translation layer (FTL), a software layer in the flash memory SSD, is maintained in a small FRAM since this metadata is a target of intensive small random writes, whereas the bulk data is kept in the flash memory. Performance evaluation based on an FPGA implementation of the Chameleon architecture shows that the use of FRAM in Chameleon improves the performance by 21.3 %. The results also show that even for bulk data that cannot be maintained in FRAM because of the size limitation, the use of fine-grained write buffering is critically important because of the inability of flash memory to perform in-placc update of data.

Predictability of earliest deadline zero laxity algorithm for multiprocessor real-time systems

Validation methods for hard real-time jobs are usually performed based on the maximum execution time. The actual execution time of jobs are assumed to be known only when the jobs arrive or not known until they finish. A predictable algorithm must guarantee that it can generate a schedule for any set of jobs such that the finish time for the actual execution time is no later than the finish time for the maximum execution time. It is known that any job-level fixed priority algorithm (such as earliest deadline first) is predictable. However, job-level dynamic priority algorithms (such as least laxity first) may or may not. In this paper, we investigate the predictability of a job-level dynamic priority algorithm EDZL (earliest deadline zero laxity). We show that EDZL is predictable on the domain of integers regardless of the knowledge of the actual execution times. Based on this result, furthermore, we also show that EDZL can successfully schedule any periodic task set if the total utilization is not greater than (m + 1)/2, where m is the number of processors

A new fair scheduling algorithm for periodic tasks on multiprocessors

We present a new scheduling algorithm, called PL that is work-conserving and in terms of schedulability, optimal on multiprocessors for a synchronous periodic task set. The PL algorithm is a laxity based algorithm and ensures execution of a task with approximate proportional fairness at each tasks period. Existing optimal algorithms on multiprocessors may cause excessive scheduling decisions and preemptions or may not be applied in a discrete environment. The proposed algorithm can be applied in a discrete environment and reduce the number of scheduling decisions and preemptions compared with a Pfair algorithm.

Power-Aware EDZL Scheduling upon Identical Multiprocessor Platforms

Upon multiprocessor platforms, global EDZL is known to be at least as effective as global EDF in scheduling task sets to meet deadlines, but there has been no research on power-aware EDZL scheduling. In this paper, we firstly address the problem of reducing energy consumption of real-time tasks on EDZL scheduling by lowering processor speed. An off-line algorithm and an on-line algorithm are proposed to reduce energy consumption while guaranteeing a hard real-time constraint. Then we show the effectiveness of our algorithms through extensive simulation.

Energy Consumption Optimization of Real-Time Embedded Systems

Minimizing energy consumption with guaranteeing real-time constraints in low-power embedded systems is gaining more importance as real-time applications become more widely used in embedded systems. Dynamic voltage scaling is a technique to reduce energy consumption by lowering supply voltage. However, lowering supply voltage may interfere with scheduling algorithms, so that tasks may not be successfully scheduled. In this paper, we formulate the problem of minimizing energy consumption for Pre-scheduling as an optimization problem, and show that the problem is a nonlinear convex optimization with linear constraints which can be solved by sequential quadratic programming. By solving the problem, we can obtain the optimal supply voltage and successful scheduling of all tasks is guaranteed.

Comparison of Tie-Breaking Policies for Real-Time Scheduling on Multiprocessor

Real-time scheduling on multiprocessor involves many complex issues. One of them is tie-breaking: the way how ties are broken among tasks with same priority. Unlike on uniprocessor, the feasibility of a task set on multiprocessor depends on a tie-breaking policy as well as on a scheduling algorithm. In this paper, we suggest and compare several tie-breaking policies for deadline-based schedule algorithms on multiprocessor. An effective tie-breaking policy for different scheduling algorithm is identified by simulation.

A Combined Approach to Improve the Response Time of Soft Aperiodic Tasks

In a real-time system with both hard deadline periodic tasks and soft deadline aperiodic tasks, hard deadline tasks must be guaranteed to meet their deadlines and soft deadline tasks are serviced in a best-effort manner. To schedule soft deadline aperiodic tasks in dynamic priority systems, one can assign a virtual deadline like the TBS algorithm or use capacity servers like the DPE algorithm. Both approaches have their own advantages, and we cannot say which one is better in terms of response time. In some cases, TBS provides shorter response time than DPE server. In other cases, the reverse may happen. In this paper, we propose a combined approach to improve the response time of soft deadline aperiodic tasks. The proposed method combines the deadline assignment and the capacity server approach. When there is enough capacity available for aperiodic tasks, the aperiodic tasks are served by capacity servers. Otherwise, our method assigns a deadline and schedule aperiodic task together with periodic tasks by EDF algorithm. In the worst case, the proposed method provides response time no later than TBS. Experimental results show that the proposed approach improves the responsiveness of soft deadline aperiodic tasks.

Improving responsiveness of soft aperiodic tasks using proportional slack time

In a real-time system with both hard real-time periodic jobs and soft real-time aperiodic jobs, it is important to guarantee that the deadline of each periodic job is met, as well as to provide a fast response time for each aperiodic job. We propose an algorithm, called Proportional Slack Reserve (PSR), that produces an efficient schedule for such an environment. For every execution unit of a periodic job, the PSR algorithm reserves time which can be used for execution of aperiodic jobs. If reserved time is not available, the algorithm assigns a deadline to an aperiodic job for achieving better responsiveness of aperiodic jobs. The proposed algorithm can fully utilize processing power while meeting all deadlines of periodic jobs. It can also easily reclaim the time unused by the periodic job. We analytically show that for each aperiodic job, the response time in a PSR schedule is no longer than that in a TBS schedule, which is known to be efficient for servicing aperiodic jobs. We also present simulation results in which the response time of PSR is significantly improved over that of TBS, and moreover the performance of PSR compares favorably with TB(N) considering scheduling overhead.

Finish Time Predictability of Earliest Deadline Zero Laxity Algorithm for Multiprocessor Real-Time Systems*This work is supported in part by Brain Korea 21 project and in part by the Korea Research Foundation Grant funded by the Korean Government (MOEHRD) (KRF-2005-041-D00636). The ICT at Seoul National University provides research facilities for this study.

This letter proves the finish time predictability of EDZL (Earliest Deadline Zero Laxity) scheduling algorithm for multiprocessor real-time systems, which is a variant of EDF. Based on the results, it also shows that EDZL can successfully schedule any periodic task set if its total utilization is not greater than (m + 1)/2, where m is the number of processors.