Improving caches in consolidated environments

Memory (cache, DRAM, and disk) is in charge of providing data and instructions to a computer's processor. In order to maximize performance, the speeds of the memory and the processor should be equal. However, using memory that always match the speed of the processor is prohibitively expensive. Computer hardware designers have managed to drastically lower the cost of the system with the use of memory caches by sacrificing some performance. A cache is a small piece of fast memory that stores popular data so it can be accessed faster. Modern computers have evolved into a hierarchy of caches, where a memory level is the cache for a larger and slower memory level immediately below it. Thus, by using caches, manufacturers are able to store terabytes of data at the cost of cheapest memory while achieving speeds close to the speed of the fastest one.^ The most important decision about managing a cache is what data to store in it. Failing to make good decisions can lead to performance overheads and over-provisioning. Surprisingly, caches choose data to store based on policies that have not changed in principle for decades. However, computing paradigms have changed radically leading to two noticeably different trends. First, caches are now consolidated across hundreds to even thousands of processes. And second, caching is being employed at new levels of the storage hierarchy due to the availability of high-performance flash-based persistent media. This brings four problems. First, as the workloads sharing a cache increase, it is more likely that they contain duplicated data. Second, consolidation creates contention for caches, and if not managed carefully, it translates to wasted space and sub-optimal performance. Third, as contented caches are shared by more workloads, administrators need to carefully estimate specific per-workload requirements across the entire memory hierarchy in order to meet per-workload performance goals. And finally, current cache write policies are unable to simultaneously provide performance and consistency guarantees for the new levels of the storage hierarchy.^ We addressed these problems by modeling their impact and by proposing solutions for each of them. First, we measured and modeled the amount of duplication at the buffer cache level and contention in real production systems. Second, we created a unified model of workload cache usage under contention to be used by administrators for provisioning, or by process schedulers to decide what processes to run together. Third, we proposed methods for removing cache duplication and to eliminate wasted space because of contention for space. And finally, we proposed a technique to improve the consistency guarantees of write-back caches while preserving their performance benefits.^

A study of branching fluid networks for enhancing the performance of thermal -fluid devices

Compact thermal-fluid systems are found in many industries from aerospace to microelectronics where a combination of small size, light weight, and high surface area to volume ratio fluid networks are necessary. These devices are typically designed with fluid networks consisting of many small parallel channels that effectively pack a large amount of heat transfer surface area in a very small volume but do so at the cost of increased pumping power requirements. ^ To offset this cost the use of a branching fluid network for the distribution of coolant within a heat sink is investigated. The goal of the branch design technique is to minimize the entropy generation associated with the combination of viscous dissipation and convection heat transfer experienced by the coolant in the heat sink while maintaining compact high heat transfer surface area to volume ratios. ^ The derivation of Murray's Law, originally developed to predict the geometry of physiological transport systems, is extended to heat sink designs which minimze entropy generation. Two heat sink designs at different scales are built, and tested experimentally and analytically. The first uses this new derivation of Murray's Law. The second uses a combination of Murray's Law and Constructal Theory. The results of the experiments were used to verify the analytical and numerical models. These models were then used to compare the performance of the heat sink with other compact high performance heat sink designs. The results showed that the techniques used to design branching fluid networks significantly improves the performance of active heat sinks. The design experience gained was then used to develop a set of geometric relations which optimize the heat transfer to pumping power ratio of a single cooling channel element. Each element can be connected together using a set of derived geometric guidelines which govern branch diameters and angles. The methodology can be used to design branching fluid networks which can fit any geometry. ^