Interplay of software bloat, hardware energy proportionality and system bottlenecks

In large flexible software systems, bloat occurs in many forms, causing excess resource utilization and resource bottlenecks. This results in lost throughput and wasted joules. However, mitigating bloat is not easy; efforts are best applied where savings would be substantial. To aid this we develop an analytical model establishing the relation between bottleneck in resources, bloat, performance and power. Analyses with the model places into perspective results from the first experimental study of the power-performance implications of bloat. In the experiments we find that while bloat reduction can provide as much as 40% energy savings, the degree of impact depends on hardware and software characteristics. We confirm predictions from our model with selected results from our experimental study. Our findings show that a software-only view is inadequate when assessing the effects of bloat. The impact of bloat on physical resource usage and power should be understood for a full systems perspective to properly deploy bloat reduction solutions and reap their power-performance benefits.

Diversity in ranking via resistive graph centers

Users can rarely reveal their information need in full detail to a search engine within 1--2 words, so search engines need to "hedge their bets" and present diverse results within the precious 10 response slots. Diversity in ranking is of much recent interest. Most existing solutions estimate the marginal utility of an item given a set of items already in the response, and then use variants of greedy set cover. Others design graphs with the items as nodes and choose diverse items based on visit rates (PageRank). Here we introduce a radically new and natural formulation of diversity as finding centers in resistive graphs. Unlike in PageRank, we do not specify the edge resistances (equivalently, conductances) and ask for node visit rates. Instead, we look for a sparse set of center nodes so that the effective conductance from the center to the rest of the graph has maximum entropy. We give a cogent semantic justification for turning PageRank thus on its head. In marked deviation from prior work, our edge resistances are learnt from training data. Inference and learning are NP-hard, but we give practical solutions. In extensive experiments with subtopic retrieval, social network search, and document summarization, our approach convincingly surpasses recently-published diversity algorithms like subtopic cover, max-marginal relevance (MMR), Grasshopper, DivRank, and SVMdiv.

Scalable multi-dimensional user intent identification using tree structured distributions

The problem of identifying user intent has received considerable attention in recent years, particularly in the context of improving the search experience via query contextualization. Intent can be characterized by multiple dimensions, which are often not observed from query words alone. Accurate identification of Intent from query words remains a challenging problem primarily because it is extremely difficult to discover these dimensions. The problem is often significantly compounded due to lack of representative training sample. We present a generic, extensible framework for learning the multi-dimensional representation of user intent from the query words. The approach models the latent relationships between facets using tree structured distribution which leads to an efficient and convergent algorithm, FastQ, for identifying the multi-faceted intent of users based on just the query words. We also incorporated WordNet to extend the system capabilities to queries which contain words that do not appear in the training data. Empirical results show that FastQ yields accurate identification of intent when compared to a gold standard.

Supervised matching of comments with news article segments

Comments constitute an important part of Web 2.0. In this paper, we consider comments on news articles. To simplify the task of relating the comment content to the article content the comments are about, we propose the idea of showing comments alongside article segments and explore automatic mapping of comments to article segments. This task is challenging because of the vocabulary mismatch between the articles and the comments. We present supervised and unsupervised techniques for aligning comments to segments the of article the comments are about. More specifically, we provide a novel formulation of supervised alignment problem using the framework of structured classification. Our experimental results show that structured classification model performs better than unsupervised matching and binary classification model.

Border basis detection is NP-complete

Border basis detection (BBD) is described as follows: given a set of generators of an ideal, decide whether that set of generators is a border basis of the ideal with respect to some order ideal. The motivation for this problem comes from a similar problem related to Grobner bases termed as Grobner basis detection (GBD) which was proposed by Gritzmann and Sturmfels (1993). GBD was shown to be NP-hard by Sturmfels and Wiegelmann (1996). In this paper, we investigate the computational complexity of BBD and show that it is NP-complete.

Extended histories: improving regularity and performance in correlation prefetchers

Data Prefetchers identify and make use of any regularity present in the history/training stream to predict future references and prefetch them into the cache. The training information used is typically the primary misses seen at a particular cache level, which is a filtered version of the accesses seen by the cache. In this work we demonstrate that extending the training information to include secondary misses and hits along with primary misses helps improve the performance of prefetchers. In addition to empirical evaluation, we use the information theoretic metric entropy, to quantify the regularity present in extended histories. Entropy measurements indicate that extended histories are more regular than the default primary miss only training stream. Entropy measurements also help corroborate our empirical findings. With extended histories, further benefits can be achieved by triggering prefetches during secondary misses also. In this paper we explore the design space of extended prefetch histories and alternative prefetch trigger points for delta correlation prefetchers. We observe that different prefetch schemes benefit to a different extent with extended histories and alternative trigger points. Also the best performing design point varies on a per-benchmark basis. To meet these requirements, we propose a simple adaptive scheme that identifies the best performing design point for a benchmark-prefetcher combination at runtime. In SPEC2000 benchmarks, using all the L2 accesses as history for prefetcher improves the performance in terms of both IPC and misses reduced over techniques that use only primary misses as history. The adaptive scheme improves the performance of CZone prefetcher over Baseline by 4.6% on an average. These performance gains are accompanied by a moderate reduction in the memory traffic requirements.

Loop transformations: convexity, pruning, and optimization

High-level loop transformations are a key instrument in mapping computational kernels to effectively exploit the resources in modern processor architectures. Nevertheless, selecting required compositions of loop transformations to achieve this remains a significantly challenging task; current compilers may be off by orders of magnitude in performance compared to hand-optimized programs. To address this fundamental challenge, we first present a convex characterization of all distinct, semantics-preserving, multidimensional affine transformations. We then bring together algebraic, algorithmic, and performance analysis results to design a tractable optimization algorithm over this highly expressive space. Our framework has been implemented and validated experimentally on a representative set of benchmarks running on state-of-the-art multi-core platforms.

Semi-supervised SVMs for classification with unknown class proportions and a small labeled dataset

In the design of practical web page classification systems one often encounters a situation in which the labeled training set is created by choosing some examples from each class; but, the class proportions in this set are not the same as those in the test distribution to which the classifier will be actually applied. The problem is made worse when the amount of training data is also small. In this paper we explore and adapt binary SVM methods that make use of unlabeled data from the test distribution, viz., Transductive SVMs (TSVMs) and expectation regularization/constraint (ER/EC) methods to deal with this situation. We empirically show that when the labeled training data is small, TSVM designed using the class ratio tuned by minimizing the loss on the labeled set yields the best performance; its performance is good even when the deviation between the class ratios of the labeled training set and the test set is quite large. When the labeled training data is sufficiently large, an unsupervised Gaussian mixture model can be used to get a very good estimate of the class ratio in the test set; also, when this estimate is used, both TSVM and EC/ER give their best possible performance, with TSVM coming out superior. The ideas in the paper can be easily extended to multi-class SVMs and MaxEnt models.

A non-syntactic approach for text sentiment classification with stopwords

The present approach uses stopwords and the gaps that oc- cur between successive stopwords –formed by contentwords– as features for sentiment classification.

Modeling the effect of transmission errors on TCP controlled transfers over infrastructure 802.11 wireless LANs

There have been several studies on the performance of TCP controlled transfers over an infrastructure IEEE 802.11 WLAN, assuming perfect channel conditions. In this paper, we develop an analytical model for the throughput of TCP controlled file transfers over the IEEE 802.11 DCF with different packet error probabilities for the stations, accounting for the effect of packet drops on the TCP window. Our analysis proceeds by combining two models: one is an extension of the usual TCP-over-DCF model for an infrastructure WLAN, where the throughput of a station depends on the probability that the head-of-the-line packet at the Access Point belongs to that station; the second is a model for the TCP window process for connections with different drop probabilities. Iterative calculations between these models yields the head-of-the-line probabilities, and then, performance measures such as the throughputs and packet failure probabilities can be derived. We find that, due to MAC layer retransmissions, packet losses are rare even with high channel error probabilities and the stations obtain fair throughputs even when some of them have packet error probabilities as high as 0.1 or 0.2. For some restricted settings we are also able to model tail-drop loss at the AP. Although involving many approximations, the model captures the system behavior quite accurately, as compared with simulations.

MAST: multi-script annotation toolkit for scenic text

This paper describes a semi-automatic tool for annotation of multi-script text from natural scene images. To our knowledge, this is the maiden tool that deals with multi-script text or arbitrary orientation. The procedure involves manual seed selection followed by a region growing process to segment each word present in the image. The threshold for region growing can be varied by the user so as to ensure pixel-accurate character segmentation. The text present in the image is tagged word-by-word. A virtual keyboard interface has also been designed for entering the ground truth in ten Indic scripts, besides English. The keyboard interface can easily be generated for any script, thereby expanding the scope of the toolkit. Optionally, each segmented word can further be labeled into its constituent characters/symbols. Polygonal masks are used to split or merge the segmented words into valid characters/symbols. The ground truth is represented by a pixel-level segmented image and a '.txt' file that contains information about the number of words in the image, word bounding boxes, script and ground truth Unicode. The toolkit, developed using MATLAB, can be used to generate ground truth and annotation for any generic document image. Thus, it is useful for researchers in the document image processing community for evaluating the performance of document analysis and recognition techniques. The multi-script annotation toolokit (MAST) is available for free download.

Does lean imply green?: a study of the power performance implications of Java runtime bloat

The presence of software bloat in large flexible software systems can hurt energy efficiency. However, identifying and mitigating bloat is fairly effort intensive. To enable such efforts to be directed where there is a substantial potential for energy savings, we investigate the impact of bloat on power consumption under different situations. We conduct the first systematic experimental study of the joint power-performance implications of bloat across a range of hardware and software configurations on modern server platforms. The study employs controlled experiments to expose different effects of a common type of Java runtime bloat, excess temporary objects, in the context of the SPECPower_ssj2008 workload. We introduce the notion of equi-performance power reduction to characterize the impact, in addition to peak power comparisons. The results show a wide variation in energy savings from bloat reduction across these configurations. Energy efficiency benefits at peak performance tend to be most pronounced when bloat affects a performance bottleneck and non-bloated resources have low energy-proportionality. Equi-performance power savings are highest when bloated resources have a high degree of energy proportionality. We develop an analytical model that establishes a general relation between resource pressure caused by bloat and its energy efficiency impact under different conditions of resource bottlenecks and energy proportionality. Applying the model to different "what-if" scenarios, we predict the impact of bloat reduction and corroborate these predictions with empirical observations. Our work shows that the prevalent software-only view of bloat is inadequate for assessing its power-performance impact and instead provides a full systems approach for reasoning about its implications.

Fast and efficient automatic memory management for GPUs using compiler-assisted runtime coherence scheme

Exploiting the performance potential of GPUs requires managing the data transfers to and from them efficiently which is an error-prone and tedious task. In this paper, we develop a software coherence mechanism to fully automate all data transfers between the CPU and GPU without any assistance from the programmer. Our mechanism uses compiler analysis to identify potential stale accesses and uses a runtime to initiate transfers as necessary. This allows us to avoid redundant transfers that are exhibited by all other existing automatic memory management proposals. We integrate our automatic memory manager into the X10 compiler and runtime, and find that it not only results in smaller and simpler programs, but also eliminates redundant memory transfers. Tested on eight programs ported from the Rodinia benchmark suite it achieves (i) a 1.06x speedup over hand-tuned manual memory management, and (ii) a 1.29x speedup over another recently proposed compiler--runtime automatic memory management system. Compared to other existing runtime-only and compiler-only proposals, it also transfers 2.2x to 13.3x less data on average.

Reconciling transactional conflicts with compiler's help

Software transactional memory(STM) is a promising programming paradigm for shared memory multithreaded programs. While STM offers the promise of being less error-prone and more programmer friendly compared to traditional lock-based synchronization, it also needs to be competitive in performance in order for it to be adopted in mainstream software. A major source of performance overheads in STM is transactional aborts. Conflict resolution and aborting a transaction typically happens at the transaction level which has the advantage that it is automatic and application agnostic. However it has a substantial disadvantage in that STM declares the entire transaction as conflicting and hence aborts it and re-executes it fully, instead of partially re-executing only those part(s) of the transaction, which have been affected due to the conflict. This "Re-execute Everything" approach has a significant adverse impact on STM performance. In order to mitigate the abort overheads, we propose a compiler aided Selective Reconciliation STM (SR-STM) scheme, wherein certain transactional conflicts can be reconciled by performing partial re-execution of the transaction. Ours is a selective hybrid approach which uses compiler analysis to identify those data accesses which are legal and profitable candidates for reconciliation and applies partial re-execution only to these candidates selectively while other conflicting data accesses are handled by the default STM approach of abort and full re-execution. We describe the compiler analysis and code transformations required for supporting selective reconciliation. We find that SR-STM is effective in reducing the transactional abort overheads by improving the performance for a set of five STAMP benchmarks by 12.58% on an average and up to 22.34%.

Multiple sub-row buffers in DRAM: unlocking performance and energy improvement opportunities

The twin demands of energy-efficiency and higher performance on DRAM are highly emphasized in multicore architectures. A variety of schemes have been proposed to address either the latency or the energy consumption of DRAMs. These schemes typically require non-trivial hardware changes and end up improving latency at the cost of energy or vice-versa. One specific DRAM performance problem in multicores is that interleaved accesses from different cores can potentially degrade row-buffer locality. In this paper, based on the temporal and spatial locality characteristics of memory accesses, we propose a reorganization of the existing single large row-buffer in a DRAM bank into multiple sub-row buffers (MSRB). This re-organization not only improves row hit rates, and hence the average memory latency, but also brings down the energy consumed by the DRAM. The first major contribution of this work is proposing such a reorganization without requiring any significant changes to the existing widely accepted DRAM specifications. Our proposed reorganization improves weighted speedup by 35.8%, 14.5% and 21.6% in quad, eight and sixteen core workloads along with a 42%, 28% and 31% reduction in DRAM energy. The proposed MSRB organization enables opportunities for the management of multiple row-buffers at the memory controller level. As the memory controller is aware of the behaviour of individual cores it allows us to implement coordinated buffer allocation schemes for different cores that take into account program behaviour. We demonstrate two such schemes, namely Fairness Oriented Allocation and Performance Oriented Allocation, which show the flexibility that memory controllers can now exploit in our MSRB organization to improve overall performance and/or fairness. Further, the MSRB organization enables additional opportunities for DRAM intra-bank parallelism and selective early precharging of the LRU row-buffer to further improve memory access latencies. These two optimizations together provide an additional 5.9% performance improvement.