Training Deep Neural Networks (DNNs) is resource-intensive and time-consuming. While prior research has explored many different ways of reducing DNN training time, the impact of input data pipeline , i.e., fetching raw data items from storage and performing data pre-processing in memory, has been relatively unexplored. This paper makes the following contributions: (1) We present the first comprehensive analysis of how the input data pipeline affects the training time of widely-used computer vision and audio Deep Neural Networks (DNNs), that typically involve complex data pre-processing. We analyze nine different models across three tasks and four datasets while varying factors such as the amount of memory, number of CPU threads, storage device, GPU generation etc on servers that are a part of a large production cluster at Microsoft. We find that in many cases, DNN training time is dominated by data stall time : time spent waiting for data to be fetched and pre-processed. (2) We build a tool, DS-Analyzer to precisely measure data stalls using a differential technique, and perform predictive what-if analysis on data stalls. (3) Finally, based on the insights from our analysis, we design and implement three simple but effective techniques in a data-loading library, CoorDL, to mitigate data stalls. Our experiments on a range of DNN tasks, models, datasets, and hardware configs show that when PyTorch uses CoorDL instead of the state-of-the-art DALI data loading library, DNN training time is reduced significantly (by as much as 5X on a single server).

Introduction:An accurate and passive fit of implant framework prosthesis, as well as the successful surgical operation is suggested as one of the critical requirements for long-term implant success.Objective:The purpose of this in vitro study was to evaluate the accuracy of the master cast using open tray impression technique with conventional and novel splinting materials.Methodology:A mandibular reference model with four ADIN implants was done. Ten custom trays were fabricated using the light curable resin sheets. Medium body polyether impression material was used. These trays were randomly divided between the two groups, with five trays in each group. Impression techniques were divided into two groups namely: Group A: Direct impression technique with open tray impression copings splinted with autopolymerizing acrylic resin (GC pattern resin). Group B: Direct impression technique with open tray impression copings splinted with Pro-temp TM 4 (bis-GMA) syringable temporization material. Thus, final impressions were made. Total of 10 master casts were fabricated. Evaluation of casts using Dynascope-Vision Engineering, TESA microhite two- dimension and coordinate measuring machine were used.Results:Statistical comparisons were made using ANOVA test and post-hoc test. Same amount of deviation values obtained with resin splinted and bis-GMA splinted impression copings.Conclusion:The master cast obtained by both the splinting material exhibits no difference from the reference model. So bis-GMA can be used, which is easy to handle, less time consuming, less technique sensitive, rigid, and readily available material in clinics.

With the arrival of the European Union's General Data Protection Regulation (GDPR), several companies are making signi cant changes to their systems to achieve compliance. The changes range from modifying privacy policies to redesigning systems which process personal data. This work analyzes the privacy policies of large-scaled cloud services which seek to be GDPR compliant. The privacy policy is the main medium of information dissemination between the data controller and the users. We show that many services that claim compliance today do not have clear and concise privacy policies. We identify several points in the privacy policies which potentially indicate non-compliance; we term these GDPR vulnerabilities. We identify GDPR vulnerabilities in ten cloud services. Based on our analysis, we propose seven best practices for crafting GDPR privacy policies.

Training Deep Neural Networks (DNNs) is resource-intensive and time-consuming. While prior research has explored many different ways of reducing DNN training time spanning different layers of the system stack, the impact of input data pipeline, i.e., fetching raw data items from storage and performing data pre-processing in memory, has been relatively unexplored.This paper makes the following contributions: (1) We present the first comprehensive analysis of how the input data pipeline affects the training time of widely-used Deep Neural Networks (DNNs). We analyze nine different models across three tasks and four datasets while varying factors such as the amount of memory, number of CPU threads, storage device, GPU generation etc on servers that are a part of a large production cluster at Microsoft. We find that in many cases, DNN training time is dominated by data stall time: time spent waiting for data to be fetched and pre-processed. ( 2) We build a tool, DS-Analyzer to precisely measure data stalls using a differential technique, and perform predictive what-if analysis on data stalls. (3) Finally, based on the insights from our analysis, we design and implement three simple but effective techniques in a data-loading library, CoorDL, to mitigate data stalls. Our experiments on a range of DNN tasks, models, datasets, and hardware configurations show that when PyTorch uses CoorDL instead of the state-of-the-art DALI data loading library, DNN training time is reduced significantly (by as much as 5× on a single server).

We present Recipe, a principled approach for converting concurrent DRAM indexes into crash-consistent indexes for persistent memory (PM). The main insight behind Recipe is that isolation provided by a certain class of concurrent in-memory indexes can be translated with small changes to crash-consistency when the same index is used in PM. We present a set of conditions that enable the identification of this class of DRAM indexes, and the actions to be taken to convert each index to be persistent. Based on these conditions and conversion actions, we modify five different DRAM indexes based on B+ trees, tries, radix trees, and hash tables to their crash-consistent PM counterparts. The effort involved in this conversion is minimal, requiring 30-200 lines of code. We evaluated the converted PM indexes on Intel DC Persistent Memory, and found that they outperform state-of-the-art, hand-crafted PM indexes in multi-threaded workloads by up-to 5.2×. For example, we built P-CLHT, our PM implementation of the CLHT hash table by modifying only 30 LOC. When running YCSB workloads, P-CLHT performs up to 2.4× better than Cacheline-Conscious Extendible Hashing (CCEH), the state-of-the-art PM hash table.