Commercial Processors Research Articles

A Survey of of Side-Channel Attacks and Mitigation for Processor Interconnects

With advancements in chip technology, the number of cores in modern commercial processors continues to rise, leading to increased complexity in interconnects and on-chip networks. This complexity, however, exposes significant security vulnerabilities, primarily in the form of side-channel and covert-channel exploits. Unlike other microarchitectural side-channel attacks, those leveraging on-chip interconnects utilize unique characteristics, allowing attackers to develop novel methods that can bypass existing effective defenses. In this paper, we present a comprehensive survey of current side-channel and covert-channel attacks based on processor-on-chip interconnects. We categorize these attacks into three types: contention-based, distance-based, and layout-based, according to the specific interconnect characteristics they exploit, and discuss corresponding countermeasures for each. Finally, we provide an outlook on future development trends in processor interconnect side channels. This survey is the first to specifically focus on interconnect-based side channels in processors.

Processing complementary foods to reduce mycotoxins in a medium scale Tanzanian mill: A hazard analysis critical control point (HACCP) approach

Designing and implementing processing procedures for producing safe complementary foods in dynamic and unregulated food systems where common food staples are frequently contaminated with mycotoxins is challenging. This paper presents lessons about minimizing aflatoxins (AF) in groundnut flour and AF and/or fumonisins (FUM) in maize and groundnut pre-blended flour for complementary feeding in the context of a dietary research intervention in rural Tanzania. The flours were processed in collaboration with Halisi Products Limited (Halisi), a medium scale enterprise with experience in milling cereal-based flours in Arusha, Tanzania. Using a hazard analysis critical control point (HACCP) approach for quality assurance, two critical control points (CCPs) for AF in processing the pre-blended flour were identified: 1) screening maize before procurement, and 2) blending during the processing of each constituent flour. Blending of maize flour was also identified as a CCP for FUM. Visual inspection during screening and sorting were identified as important control measures for reducing AF, but these steps did not meet the criteria for a CCP due to lack of objective measurement and verifiable standards for AF. The HACCP approach enabled the production of low AF (<5 μg/kg) and FUM (<2 μg/g) flours with low rejection rates for the final products. The paper presents practical lessons that could be of value to a range of commercial processors in similar low- and middle-income contexts who are keen on improving food quality.

PSV-23 Characterization of Katahdin Lamb Carcasses in a Traditional Lamb Marketing Context

Abstract The Katahdin breed has grown in popularity in the US, but it is relatively unknown if their carcasses conform to expectations of the traditional commodity lamb industry. Purebred Katahdin wethers (n = 223) were sourced from the genetic reference flock at the U.S. Meat Animal Research Center over two consecutive years. The flock is enrolled in the National Sheep Improvement Program (NSIP) and estimated breeding values (EBV) were available on standard production traits. Lambs were assigned to one of three groups per year based on body weight (BW) and fed a common finishing diet through harvest at similar end BW (51.3 ± 5.6 kg) but varying age (223.8 ± 32.2 d). Lambs were harvested at a commercial processor, hot carcass weight (HCW) was recorded, dressing percentage (DP) was calculated, and carcasses were chilled for 24 h. Carcasses were then ribbed and 12th rib backfat thickness (BFT), and loin muscle area (LMA) were measured. Traits were analyzed in mixed models with fixed class effects of finishing group (6 levels) and rearing type (single or multiple), a linear covariate of harvest age, and a random effect of sire (n = 16). There were no differences in BW, HCW, BFT, and LMA between rearing types but lambs reared as multiples had greater DP than those reared as singles (P = 0.02; Table). End BW increased while DP decreased with lamb harvest age (P ≤ 0.03). Pearson correlation coefficients were estimated between traits including post-weaning weight (PWWT) NSIP EBV. There was a strong, positive correlation between HCW and BFT (0.52) and HCW and BWT (0.54; P &amp;lt; 0.001). In contrast, HCW was only moderately correlated to LMA (0.35; P &amp;lt; 0.001). The PWWT EBV is the best genetic predictor of post-weaning lamb growth and was moderately correlated with end BW (0.38), HCW (0.35), and LMA (0.35; P &amp;lt; 0.001). Traditional markets prefer carcasses to be 29.5 – 36.3 kg HCW and &amp;lt; 6.4 mm BFT; 91% and 51% of Katahdin carcasses were below the HCW and BFT thresholds, respectively. Katahdins are capable of producing lean carcasses that conform to packer ideals for BF when fed to harvest BW of the present study, which may reduce excess fat trimming during fabrication. However, HCW observed in the present study may suggest greater suitability for non-traditional (i.e., ethnic) markets. Lower correlations between carcass traits in Katahdin compared with past research in Western white-faced lambs may highlight important biological differences in growth and fat deposition among divergent breed types. Crossbreeding Katahdin ewes with terminal sire breeds and within breed selection will improve growth and carcass traits without exceeding HCW thresholds commonly reached in wool sheep but should be considered in tandem with other breeding objectives.

27 Quality Variation in Beef Chuck Eye Rolls Sourced from a Commercial Processor

Abstract The objective of this study was to evaluate quality variation in beef Chuck Eye rolls (IMPS 116D). Beef Chuck Rolls (IMPS 116A) were purchased from a commercial meat packer and fabricated to obtain beef Chuck Eye Rolls (IMPS 116D). Cases were transported to the Illinois State University Meat Lab and hand-sliced into 2.54 cm boneless beef Chuck Eye steaks (IMPS 1116D) (n = 161). Each roll contained between 7 to 10 steaks for use in further meat quality analyses in accordance with in-plant procedures. Upon slicing, steaks were allowed to bloom for 15 minutes and then used in subsequent quality analysis including color, cook loss, and tenderness evaluation. Instrumental color measurements [L* (lightness), a* (redness), b* (yellowness)] were taken for each steak in duplicate with a HunterLab Miniscan XE Plus Spectrophotometer using a D65 illuminant, 100 observer, and 35-mm aperture. All steaks were cooked to a final internal temperature of 71° C. Steaks then underwent cook loss evaluation as well as tenderness analysis via Warner-Bratzler shear force (WBSF). Statistical analysis was performed using the Proc MIXED procedure of SAS to estimate variance components of steak quality in a nested design (steak within roll within case). For all response variables, the standard error was large for the case term. Ranges of observed response values were: L* 28.03-52.15, a* 14.55-27.82, b* 12.78-25.32, cook loss 17.58-61.11, shearforce 2.10-5.86. Variance components in L* and a* values followed similar patterns in which the greatest amount of variance was contributed by steak, followed by case, followed by roll (L* 13.541 (P &amp;lt; 0.001) vs. 9.035 (P = 0.1790) vs. 1.958 (P = 0.1839) and a* 4.055 (P &amp;lt; 0.0001) vs. 1.605 ( P= 0.2124) vs. 0.632 (P = 0.1746). Case, roll, and steak contributed variance almost equally for b* though two variance components were statistically different from zero σ2case =2.133 (P = 0.2344) σ2roll = 1.627 (P = 0.0335) σ2steak = 2.37 (P &amp;lt; 0.0001). Case contributed approximately 1.5 times the variance in cook loss when compared with steak (σ2case = 43.687 (P = 0.1387) vs. σ2steak=29.471 (P &amp;lt; 0.0001)) and approximately 3.5 times more than roll ( σ2case = 43.687 vs. σ2roll = 12.077 (P = 0.0565)). The largest impact on variance in shearforce values came from steak whereas case and roll contributed almost equally (shearforce σ2case = 0.0256 (P = 0.3894)σ2roll = 0.0333 (P = 0.2040) σ2steak = 0.2454 (P &amp;lt; 0.0001)). Some correlations were observed: L*:a* (-0.46, P &amp;lt; 0.0001), L*:b* (-0.07, P = 0.3979), L*:cookloss (0.166, P = 0.0422), and cookloss:shearforce (0.274, P = 0.0007). As lower-cost alternatives to traditional center of the plate cuts continue to grow in popularity among large volume purchasers, it is important to evaluate potential differences in quality attributes. Further evaluation of the impact of case as well as steak within Chuck Eye roll on individual steak quality is warranted.

Radix-64 Floating-Point Division and Square Root: Iterative and Pipelined Units

Digit-recurrence algorithms are widely used in actual microprocessors to compute floating-point division and square root. These iterative algorithms present a good trade-off in terms of performance, area and power. Traditionally, commercial processors have iterative division and square root units where the iteration logic is used over several cycles. The main drawbacks of these iterative units are long latency and low throughput due to the reuse of part of the logic over several cycles, and its hardware complexity with separated logic for division and square root. We present a radix-64 floating-point division and square root algorithm with a common iteration for division and square root and where, to have an affordable implementation, each radix-64 iteration is made of two simpler radix-8 iterations. The radix-64 algorithm allows to get low-latency operations, and the common division and square root radix-64 iteration results in some area reduction. The algorithm is mapped into two different microarchitectures: a low-latency and low area iterative unit, and a low-latency and high-throughput pipelined unit. In both units speculation between consecutive radix-8 iterations is used to reduce the timing.

AVX-TSCHA: Leaking information through AVX extensions in commercial processors

Modern x86 processors support an AVX instruction set to boost performance. However, this extension set may also cause security issues. We discovered that there are vulnerable properties in the implementation of the masked load/store instructions. First, these instructions can suppress exceptions caused by invalid or inaccessible memory access. Second, the execution time of these instructions leaks the current state of the page mappings, permissions, and TLB states.Based on this, we present a novel AVX timing side-channel attack that can defeat address space layout randomization. We demonstrate the significance of our side-channel attack by showing User and Kernel ASLR breaks on the recent Intel and AMD processors in various environments, including cloud computing systems (Amazon AWS, Google GCP, and Microsoft Azure), an SGX enclave (a fine-grained ASLR break), and major OSes (Linux, Windows, and macOS). Our attack can identify the Linux kernel's base address in 0.29 ms as well as those of loaded kernel modules in 2.24 ms, with a near-zero error rate. We further demonstrate that our attack can be used to infer user behavior, such as mouse movements and data transmissions over the network. Our evaluation results on multiple mobile, desktop, and server processors (a total of 26 Intel and AMD CPUs) show that 1) the AVX timing side-channel works on the vast majority of Intel processors (from the Sandy Bridge microarchitecture) as well as AMD processors (from the Zen microarchitecture onward) and 2) our KASLR breaks are very fast and reliable. To the best of our knowledge, our attack is the first to demonstrate a KASLR break on both the recent Intel Alder Lake and AMD Zen 3 CPUs. We highlight that more robust isolation or fine-grained randomization should be adopted to mitigate our presented attacks successfully.

A Comprehensive Modeling Platform for Interconnect Technologies

With device scaling facing major physical and manufacturing challenges, many research efforts have been focused on the impact of device parameters and designs on circuit performance and power dissipation. However, similar research on the impact of Cu-low <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">k</i> interconnects at the advanced nodes and the potential benefits offered by novel wire and via material and process options is lacking. While studies based on smaller circuit blocks exist, the exploration of these benefits for the performance of larger processors is lacking although the limits imposed by interconnect resistances become more severe due to longer wires and increased complexity in larger circuits. In this article, we conduct a study to quantify the effect of back-end-of-the-line (BEOL) technology improvements on circuit power, performance, and area (PPA) of various circuits from small blocks up to a full commercial processor with L1-cache. We modified the interconnect technology (ICT) files for the ASAP7 process design kit (PDK) to reflect various BEOL options such as thinner barrier/liners, near barrierless vias, and alternative material options based on TCAD simulations. We show that these options can offer up to a 1.769 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\times$</tex-math> </inline-formula> improvement in effective frequency or up to three metal-level savings.

Effects of the border closure Benin-Nigeria on the marketing of Clarias gariepinus

The Clarias is consumed in Africa and Benin. Despite its important role in food, its sector is subject to many constraints, including marketing in Nigeria. The objective of this study is to assess the marketing of smoked Clarias in southern Benin in the face of the Nigerian border closure. It is made in the Oueme department. Individual surveys based on structured questionnaires were conducted with 30 fish farmers and 45 processors. From the results obtained, three Nigerian markets are known for the sale of smoked Clarias, of which Badagri is the main destination. The actors in this activity are mostly adults with an average experience over 20 years. Fish farming is mainly the prerogative of men (83.33%) while processing and marketing are the only privilege of women (100%). The closure of the Nigerian border has led to a decrease in the annual fish production of the respondents from 8,960 kg to 7,491.96 kg as well as a decrease in the quantity of smoked fish from 6,249.24 kg to 2,319.96 kg after the border closure. The study shows a decline in the level of income of fish farmers and commercial processors, especially in the achievement of food security in their households. The marketing of smoked fish in Nigeria appears to be a profitable activity for processors.

Rework Practices Used During Milk Processing: An Industry Survey

Rework is a common practice in the dairy industry for processors to minimize waste while recovering costs from products that are unsaleable. Regulations related to reworking fluid dairy products are focused on product safety; however, rework in the fluid milk industry and its implications for product quality have not been previously investigated. Our objectives were to characterize current industry practices for reworking fluid dairy products and identify scenarios that could contribute to reduced product quality, particularly microbial spoilage. Seven commercial fluid milk processors from the Pacific Northwest were interviewed regarding their rework handling practices. Processors used various terms (rework, reclaim, and rerun) to describe specific product recovery, storage, and reprocessing procedures. Processors reported nine typical rework motivations, with reclaim and packaging problems the most common; however, rework also played an important role in handling special circumstances. Milk products were reworked as soon as 3 days after production up to the code date (21 days) at dilution rates of ≤20% rework to ≥80% fresh product. Rework conditions with the potential to influence product quality or shelf life of milk products were identified.

BTB-X: A Storage-Effective BTB Organization

Many contemporary applications feature multi-megabyte instruction footprints that overwhelm the capacity of branch target buffers (BTB) and instruction caches (L1-I), causing frequent front-end stalls that inevitably hurt performance. BTB is crucial for performance as it enables the front-end to accurately resolve the upcoming execution path and steer instruction fetch appropriately. Moreover, it also enables highly effective fetch-directed instruction prefetching that can eliminate many L1-I misses. For these reasons, commercial processors allocate vast amounts of storage capacity to BTBs. This letter aims to reduce BTB storage requirements by optimizing the organization of BTB entries. Our key insight is that today's BTBs store the full target address for each branch, yet the vast majority of dynamic branches have short offsets requiring just a handful of bits to encode. Based on this insight, we organize the BTB as an ensemble of smaller BTBs, each storing offsets within a particular range. Doing so enables a dramatic reduction in storage for target addresses. We also compress tags to reduce the tag storage cost. Our final design, called BTB-X, uses an ensemble of five BTBs with compressed tags that enables it to track 2.8x more branches than a conventional BTB with the same storage budget.

Correlation of Bioelectrical Impedance With Freshness Quality Attributes of Beef Longissimus Lumborum Steaks

The quality attributes of beef longissimus lumborum during 15 d of retail display were assessed using surface bioelectrical impedance analysis (S-BIA) and internal bioelectrical impedance analysis (I-BIA). Beef loins (N = 18) were obtained from 3 commercial processors with 3 postmortem (PM) ages (27, 34, and 37 d). Loins were fabricated into twelve 2.54-cm-thick steaks, subdivided into 6 consecutively cut pairs, and randomly assigned to one of 6 display days (DD): 0, 3, 6, 9, 12, or 15. Steaks were assessed for S-BIA and I-BIA. Three locations were analyzed within each steak: top, middle, and bottom. Microbiological analysis, BIA, pH, instrumental color, proximate composition, and lipid oxidation were measured. There was a location × PM day × DD interaction (P &amp;lt; 0.05) for longissimus lumborum steaks for S-BIA. Among all 3 locations, steaks aged 27 d had higher (P &amp;lt; 0.05) S-BIA values on day 9 and 12 than steaks aged 34 and 37 d. There were no location × PM day × DD or two-way interactions (P &amp;gt; 0.05) for I-BIA. Display day affected (P &amp;lt; 0.05) all instrumental color data regardless of PM aging times. Among all PM aging times, steaks aged 27 d were 13% and 7% higher for a* and b* , respectively, compared with 34 and 37 d PM. There was a PM day × DD interaction (P &amp;lt; 0.05) for aerobic plate counts (APC). From day 0 and 9 of display, APC of steaks aged 27 d PM were 1 to 2.0 log colony-forming units/cm 2 lower than steaks aged 34 and 37 d. Quality attributes, including a*, b* , APC, and thiobarbituric acid reactive substances, were correlated (r = 0.70, − 0.64, − 0.56, and 0.69, respectively) with S-BIA. Overall, BIA values increased on aerobically packaged longissimus lumborum steaks and were correlated with various freshness quality parameters.

An Embedded Quaternion-Based Extended Kalman Filter Pose Estimation for Six Degrees of Freedom Systems

This paper proposes a formulation of quaternion-based Extended Kalman Filter pose estimation for six degrees of freedom systems embedded in an FPGA with commercial processors. Our approach uses the fusion of a camera and an inertial measurement unit to estimate simultaneously the position and the orientation of the system of interest. In addition, a Stewart platform is used to validate and evaluate the estimated pose. Although this work considers the use of common low-cost sensors and the use of markers with simple geometry, the results show excellent performance of the developed filter, being able to estimate the pose and orientation with an error below 8.14 mm and 0.63o, respectively. Furthermore, the effectiveness of the approach has also been evaluated, showing that the filter is able to converge quickly when the markers are retrieved after a loss of camera data for a short period of time.

Impact of glyphosate applied preharvest on oat kernel quality

AbstractBackground and objectivesPreharvest glyphosate is applied to oats for weed control and as a desiccant. The effect of glyphosate applied at various stages of oat physiological maturity on oat end product quality has not been reported previously. In this study, the effects of glyphosate application at two separate stages of physiological maturity on oat groat quality and rolled oats quality were determined.FindingsOat was grown at two locations. Glyphosate was applied at the soft dough, physiological maturity stages, or not applied. Effect of preharvest glyphosate application at the oat soft dough stage on groat hardness, groat percentage (0.7%–3.7% decrease by cultivar), and percent plump groats (0.3%–6.1% decrease by cultivar), when compared to untreated oats, was noticeable. However, glyphosate applied at physiological maturity (manufacturer's recommendation) did not detrimentally affect groat starch (average 60.3%) or end product quality.ConclusionsTherefore, from an end‐user perspective, applying glyphosate preharvest at oat physiological maturity does not impact oat quality.Significance and noveltyGlyphosate when applied preharvest to oats can control weed and facilitate harvest, a boon to farmers. Final oat quality, a decisive factor for commercial processor use, was not detrimentally affected when glyphosate was applied at oat physiological maturity.

Leaking Information Through Cache LRU States in Commercial Processors and Secure Caches

The Least-Recently Used (LRU) cache replacement policy and its variants are widely deployed in modern processors. This article shows in detail that the LRU states of caches can be used to leak information: any access to a cache by a sender will modify the LRU state, and the receiver is able to observe this through a timing measurement. This article presents LRU timing-based channels both when the sender and the receiver have access to shared memory, e.g., shared library, and when they are separate processes without shared memory. In addition, the new LRU timing-based channels are demonstrated on both Intel and AMD processors in scenarios where the sender and the receiver are sharing the cache in both hyper-threaded setting and time-sliced setting. The transmission rates of the LRU channels can be up to 600 Kbps per cache set in the hyper-threaded setting. Different from the majority of existing cache channels which require the sender to trigger cache misses, the new LRU channels work with the sender only having cache hits, making the channel faster and stealthier. This article further discusses the effectiveness of the new LRU channels against a number of secure cache designs. Especially, the LRU channels are demonstrated to work against two representative secure caches, Partition-Locked (PL) cache and Random Fill (RF) cache, in the gem5 simulator, showing possible vulnerabilities in the secure cache designs in which the security of the replacement state is not protected properly.

Supply and Value Chain Analysis of Freshwater Sardine, Sardinella tawilis (Herre 1927), in Taal Lake, Batangas, Philippines

The supply and value chain of the world’s only freshwater sardine, Sardinella tawilis endemic to Taal Lake, were studied from January to December 2016. This study aimed to identify the actors in the value chain, evaluate each actor’s value addition, identify the roles of men and women in the chain, and identify the issues, concerns, and entry points for intervention. Key informant interviews, focus group discussions, and tracer survey interviews were done to gather data. A semi-structured questionnaire was directed to 189 respondents within and outside Taal Lake. The study showed that the tawilis marketing system is limited to the local market. Tawilis is traded fresh and processed. The chain’s key actors include the fishers, fish buyers (wholesaler, retailer, peddlers, and contracted fish buyers), processors, and consumers. The outcome of the value chain analysis of the tawilis industry showed that commercial processors have the highest value-added due to the place, form, and time transformation of the product. Meanwhile, the fishers and small-scale fish buyers have the lowest value-added during lean and peak season, respectively. The tawilis industry provides livelihood to the marginal fisherfolk, which is the first supply chain link. Several strategies were recommended in the form of process, product, function, and overall upgrading to uplift the economic benefit of the different actors in the chain and boost the tawilis industry. These include the improvement on the fishing operations, upgrading of fishing gear and other paraphernalia, provision of training on post-harvest techniques (handling, preservation, processing, value-adding, product development, etc.), market matching strategies, improvement in farm to market road transportation, establishment of fish processing facilities, and access to credit, loans or grants from the national and local governments.

DeTraS: Delaying Stores for Friendly-Fire Mitigation in Hardware Transactional Memory

Commercial Hardware Transactional Memory (HTM) systems are best-effort designs that leverage the coherence substrate to detect conflicts eagerly. Resolving conflicts in favor of the requesting core is the simplest option for ensuring deadlock freedom, yet it is prone to livelocks. In this work, we propose and evaluate DeTraS (Delayed Transactional Stores), an HTM-aware store buffer design aimed at mitigating such livelocks. DeTraS takes advantage of the fact that modern commercial processors implement a large store buffer, and uses it to prevent transactional stores predicted to conflict from performing early in the transaction. By leveraging existing processor structures, we propose a simple design that improves the ability of requester-wins HTM systems to achieve forward progress in spite of high contention while side-stepping the performance penalty of falling back to mutual exclusion. With just over 50 extra bytes, DeTraS captures the advantages of lazy conflict management without the complexity brought into the coherence fabric by commit arbitration schemes nor the relaxation of the single-writer invariant of prior works. Through detailed simulations of a 16-core tiled CMP using gem5, we demonstrate that DeTraS brings reductions in average execution time of 25 percent when compared to an Intel RTM-like design.

41 Characterizing the amount and variability of intramuscular fat deposition throughout the loin using barrows and gilts from two genotypes

Abstract Pork quality is evaluated by commercial processors along the ventral surface of whole boneless loins. However, in retail settings boneless chops are evaluated individually for quality. The objective was to determine the variability in marbling attributable to chop location, sex, and sire line. The population of pigs evaluated were sired by commercial duroc boars selected for meat quality (MQ; n=96) or lean growth (LG; n=96) and equally split between sexes. Upon slaughter and fabrication, bone-in chops were removed from four locations along the loin (A=6th rib, B=10th rib, C=last rib, D=4th lumbar). Pairs of chops from each location were used for visual color, visual marbling, and subjective firmness, Warner-Bratzler shear force, and determination of moisture and extractable lipid (IMF). Variability across location, sex, and sire line was measured using a Levene’s test and the VARCOMP procedure of SAS was used to evaluate the proportion of variability each contributed to total variance. Barrows produced chops with greater IMF (3.64%) and variability (P&amp;lt; 0.001) than gilts (3.20%; P&amp;lt; 0.001). Chops from MQ pigs exhibited greater IMF (4.02%) and variability (P&amp;lt; 0.001) than LG (2.82%; P&amp;lt; 0.001). Chops from location A (3.80%) and D (3.77%) had greater IMF than B (3.34%; P&amp;lt; 0.001) with all three locations greater than C (2.77%; P&amp;lt; 0.001). Variance (s2) also differed (A=1.44% B=1.59% C=1.05% D=2.18%; P&amp;lt; 0.05) across chop locations. Of the variability in IMF, 33.0% was attributed to sire line, 10.16% to chop location and 4.01% to sex, with 52.83% unaccounted for. Location A chops were the most tender (2.57kg; P&amp;lt; 0.001) and C chops the least tender (2.93kg; P&amp;lt; 0.01). No differences in variability (s2; P=0.40) of tenderness were observed across chop location (A=0.31kg B=0.24kg C=0.24kg D=0.23kg). In conclusion, chop location, sex, and sire line all contribute to the amount and variability of pork loin marbling. Supported by National Pork Board Grant #18-118.

Phase-Aware Cache Partitioning to Target Both Turnaround Time and System Performance

The Last Level Cache (LLC) plays a key role in the system performance of current multi-cores by reducing the number of long latency main memory accesses. The inter-application interference at this shared resource, however, can lead the system to undesired situations regarding performance and fairness. Recent approaches have successfully addressed fairness and turnaround time (TT) in commercial processors. Nevertheless, these approaches must face sustaining system performance, which is challenging. This work makes two main contributions. LLC behaviors regarding cache performance, data reuse and cache occupancy, that adversely impact on the final performance are identified. Second, based on these behaviors, we propose the Critical-Phase Aware Partitioning Approach (CPA), which reduces TT while sustaining (and even improving) IPC by making an effective use of the LLC space. Experimental results show that CPA outperforms CA, Dunn and KPart state-of-the-art approaches, and improves TT (over 40 percent in some workloads) over Linux default behavior while sustaining or even improving IPC by more than 3 percent in several mixes.

Effect of Pre-Blended Phosphates on the Freezing Quality Characteristics of Ground Woody Breast Meat Compared to Normal Meat.

Simple SummaryWoody breast (WB) myopathy is a major meat quality issue facing the global poultry industry. The WB affected meat can be ground, frozen, and utilized for further processed poultry products as needed. Freezing and thawing of WB meat may affect the functional properties of myofibrillar proteins such as lower water holding capacity and gelation. We found that pre-blended phosphate can increase the functionality of freshly ground WB meat and maintain it after freezing. We have demonstrated that the addition of phosphates can help the utilization of ground woody breast meat even after frozen storage.Woody breast (WB) myopathy affected meat has a tough texture, higher cook loss, and decreased water holding capacity (WHC), and thus lower consumer acceptability. The WB meat can be ground and further converted into further processed products or frozen, stored, and shipped to further processors. Freezing and thawing of ground WB meat may further affect the quality of WB meat products. Hence, research is required to determine the effect of pre-blended phosphates on the quality of ground WB meat as well as its cryoprotective effect during frozen storage. The objective of this experiment was to investigate the effect of pre-blended phosphate levels on meat quality in WB and normal breast (NB) fillets before and after freezing. NB fillets and severely affected WB fillets were procured from a local commercial processor. The meat was separated into various treatment groups according to the sodium tripolyphosphate (STPP) inclusion levels (0, 0.25, and 0.5% w/w). The meat was ground with respective phosphate treatments and subdivided into vacuum-sealed bags (n = 240; 1 kg each). Half of the bags (n = 120) from each treatment were taken for meat quality analysis, while the other bags were placed in a freezer (−18 °C) for 6 days. Fresh samples were analyzed within 6–8 h while the frozen samples were thawed for 18 h at 4°C prior to analysis. Samples (n = 10) were analyzed for gel strength, pH, color (L* a* b*), proximate composition, and randomly selected samples (n = 5) were analyzed for aerobic plate count (APC). Experiments were repeated in two separate replications and the data was analyzed using the Proc Glimmix model procedure in SAS (v. 9.4) (Cary, NC, USA) with LSMeans Separation at p ≤ 0.05. The gel strength (g) of the fresh ground NB meat (883.7 g) was higher than the gel strength of woody meat (720.8 g) with 0% phosphate (p ≤ 0.05). Addition of phosphate (0.25 and 0.5%) significantly increased the gel strength of fresh woody meat but it was significantly lower than NB meat added with 0.25 and 0.5% phosphate treatment. After freezing, ground NB meat samples with 0.25 and 0.5% phosphate had higher gel strength compared to fresh and frozen ground WB meat (p ≤ 0.05). Pre-blended STPP raised the pH in all treatments (p < 0.05). Treatments did not have any clear impact on APC of ground WB or NB meat. Addition of pre-blended sodium tripolyphosphate increases the functionality of fresh and frozen ground WB meat, as well as NB meat.

Accuracy-Power Controllable LiDAR Sensor System with 3D Object Recognition for Autonomous Vehicle.

Light detection and ranging (LiDAR) sensors help autonomous vehicles detect the surrounding environment and the exact distance to an object’s position. Conventional LiDAR sensors require a certain amount of power consumption because they detect objects by transmitting lasers at a regular interval according to a horizontal angular resolution (HAR). However, because the LiDAR sensors, which continuously consume power inefficiently, have a fatal effect on autonomous and electric vehicles using battery power, power consumption efficiency needs to be improved. In this paper, we propose algorithms to improve the inefficient power consumption of conventional LiDAR sensors, and efficiently reduce power consumption in two ways: (a) controlling the HAR to vary the laser transmission period () of a laser diode (LD) depending on the vehicle’s speed and (b) reducing the static power consumption using a sleep mode, depending on the surrounding environment. The proposed LiDAR sensor with the HAR control algorithm reduces the power consumption of the LD by 6.92% to 32.43% depending on the vehicle’s speed, compared to the maximum number of laser transmissions (). The sleep mode with a surrounding environment-sensing algorithm reduces the power consumption by 61.09%. The algorithm of the proposed LiDAR sensor was tested on a commercial processor chip, and the integrated processor was designed as an IC using the Global Foundries 55 nm CMOS process.