Improving Superscalar Instruction Dispatch And Issue By Exploiting Dynamic Code Sequences

Superscalar processors currently have the potential to fetch multiple basic blocks per cycle by employing one of several recently proposed instruction fetch mechanisms. However, this increased fetch bandwidth cannot be exploited unless pipeline stages further downstream correspondingly improve. In particular,register renaming a large number of instructions per cycle is diDcult. A large instruction window, needed to receive multiple basic blocks per cycle, will slow down dependence resolution and instruction issue. This paper addresses these and related issues by proposing (i) partitioning of the instruction window into multiple blocks, each holding a dynamic code sequence; (ii) logical partitioning of the registerjle into a global file and several local jles, the latter holding registers local to a dynamic code sequence; (iii) the dynamic recording and reuse of register renaming information for registers local to a dynamic code sequence. Performance studies show these mechanisms improve performance over traditional superscalar processors by factors ranging from 1.5 to a little over 3 for the SPEC Integer programs. Next, it is observed that several of the loops in the benchmarks display vector-like behavior during execution, even if the static loop bodies are likely complex for compile-time vectorization. A dynamic loop vectorization mechanism that builds on top of the above mechanisms is briefly outlined. The mechanism vectorizes up to 60% of the dynamic instructions for some programs, albeit the average number of iterations per loop is quite small.

Enhancing Speedup in Network Processing Applications by Exploiting Instruction Reuse with Flow Aggregation

Instruction reuse is a microarchitectural technique that improves the execution time of a program by removing redundant computations at run-time. Although this is the job of an optimizing compiler, they do not succeed many a time due to limited knowledge of run-time data. In this paper we examine instruction reuse of integer ALU and load instructions in network processing applications. Specifically, this paper attempts to answer the following questions: (1) How much of instruction reuse is inherent in network processing applications?, (2) Can reuse be improved by reducing interference in the reuse buffer?, (3) What characteristics of network applications can be exploited to improve reuse?, and (4) What is the effect of reuse on resource contention and memory accesses? We propose an aggregation scheme that combines the high-level concept of network traffic i.e. "flows" with a low level microarchitectural feature of programs i.e. repetition of instructions and data along with an architecture that exploits temporal locality in incoming packet data to improve reuse. We find that for the benchmarks considered, 1% to 50% of instructions are reused while the speedup achieved varies between 1% and 24%. As a side effect, instruction reuse reduces memory traffic and can therefore be considered as a scheme for low power.

Relative roles of instruction count and cycles per instruction in WCET estimation

Most of the existing WCET estimation methods directly estimate execution time, ET, in cycles. We propose to study ET as a product of two factors, ET = IC * CPI, where IC is instruction count and CPI is cycles per instruction. Considering directly the estimation of ET may lead to a highly pessimistic estimate since implicitly these methods may be using worst case IC and worst case CPI. We hypothesize that there exists a functional relationship between CPI and IC such that CPI=f(IC). This is ascertained by computing the covariance matrix and studying the scatter plots of CPI versus IC. IC and CPI values are obtained by running benchmarks with a large number of inputs using the cycle accurate architectural simulator, Simplescalar on two different architectures. It is shown that the benchmarks can be grouped into different classes based on the CPI versus IC relationship. For some benchmarks like FFT, FIR etc., both IC and CPI are almost a constant irrespective of the input. There are other benchmarks that exhibit a direct or an inverse relationship between CPI and IC. In such a case, one can predict CPI for a given IC as CPI=f(IC). We derive the theoretical worst case IC for a program, denoted as SWIC, using integer linear programming(ILP) and estimate WCET as SWIC*f(SWIC). However, if CPI decreases sharply with IC then measured maximum cycles is observed to be a better estimate. For certain other benchmarks, it is observed that the CPI versus IC relationship is either random or CPI remains constant with varying IC. In such cases, WCET is estimated as the product of SWIC and measured maximum CPI. It is observed that use of the proposed method results in tighter WCET estimates than Chronos, a static WCET analyzer, for most benchmarks for the two architectures considered in this paper.

Annexin A2 and PSF proteins interact with p53 IRES and regulate translation of p53 mRNA

p53 mRNA has been shown to be translated into two isoforms, full-length p53 (FL-p53) and a truncated isoform Delta N-p53, which modulates the functions of FL-p53 and also has independent functions. Previously, we have shown that translation of p53 and Delta N-p53 can be initiated at Internal Ribosome Entry Sites (IRES). These two IRESs were shown to regulate the translation of p53 and Delta N-p53 in a distinct cell-cycle phase-dependent manner. Earlier observations from our laboratory also suggest that the structural integrity of the p53 RNA is critical for IRES function and is compromised by mutations that affect the structure as well as RNA protein interactions. In the current study, using RNA affinity approach we have identified Annexin A2 and PTB associated Splicing Factor (PSF/SFPQ) as novel ITAFs for p53 IRESs. We have showed that the purified Annexin A2 and PSF proteins specifically bind to p53 IRES elements. Interestingly, in the presence of calcium ions Annexin A2 showed increased binding with p53 IRES. Immunopulldown experiments suggest that these two proteins associate with p53 mRNA ex vivo as well. Partial knockdown of Annexin A2 and PSF showed decrease in p53 IRES activity and reduced levels of both the p53 isoforms. More importantly the interplay between Annexin A2, PSF and PTB proteins for binding to p53mRNA appears to play a crucial role in IRES function. Taken together, our observations suggest pivotal role of two new trans-acting factors in regulating the p53-IRES function, which in turn influences the synthesis of p53 isoforms.

Distinctive contributions of the ribosomal P-site elements m(2)G966, m(5)C967 and the C-terminal tail of the S9 protein in the fidelity of initiation of translation in Escherichia coli

The accuracy of pairing of the anticodon of the initiator tRNA (tRNA(fMet)) and the initiation codon of an mRNA, in the ribosomal P-site, is crucial for determining the translational reading frame. However, a direct role of any ribosomal element(s) in scrutinizing this pairing is unknown. The P-site elements, m(2)G966 (methylated by RsmD), m(5)C967 (methylated by RsmB) and the C-terminal tail of the protein S9 lie in the vicinity of tRNA(fMet). We investigated the role of these elements in initiation from various codons, namely, AUG, GUG, UUG, CUG, AUA, AUU, AUC and ACG with tRNA(CAU)(fmet) (tRNA(fMet) with CAU anticodon); CAC and CAU with tRNA(GUG)(fme); UAG with tRNA(GAU)(fMet) using in vivo and computational methods. Although RsmB deficiency did not impact initiation from most codons, RsmD deficiency increased initiation from AUA, CAC and CAU (2- to 3.6-fold). Deletion of the S9 C-terminal tail resulted in poorer initiation from UUG, GUG and CUG, but in increased initiation from CAC, CAU and UAC codons (up to 4-fold). Also, the S9 tail suppressed initiation with tRNA(CAU)(fMet)lacking the 3GC base pairs in the anticodon stem. These observations suggest distinctive roles of 966/967 methylations and the S9 tail in initiation.

Integrated relative translation and rotation control of spacecraft formations

Formation flying of small spacecraft provides a way to improve the resolution by aperture distribution. This requires autonomous control of relative position and relative attitude. The present work addresses the formation control using a PID controller to maintain both relative position and relative attitude. To avoid continuous pulsing due to noise, a dead-band has been provided in the position loop. PID control has been selected to maintain the formation in the presence of unmodeled disturbances. Simulations show that the proposed controller meets the required translational and rotational relative motions even in the presence of disturbances.

The translation initiation factor DAP5 promotes IRES-driven translation of p53 mRNA

Translational regulation of the p53 mRNA can determine the ratio between p53 and its N-terminal truncated isoforms and therefore has a significant role in determining p53-regulated signaling pathways. Although its importance in cell fate decisions has been demonstrated repeatedly, little is known about the regulatory mechanisms that determine this ratio. Two internal ribosome entry sites (IRESs) residing within the 5'UTR and the coding sequence of p53 mRNA drive the translation of full-length p53 and Delta 40p53 isoform, respectively. Here, we report that DAP5, a translation initiation factor shown to positively regulate the translation of various IRES containing mRNAs, promotes IRES-driven translation of p53 mRNA. Upon DAP5 depletion, p53 and Delta 40p53 protein levels were decreased, with a greater effect on the N-terminal truncated isoform. Functional analysis using bicistronic vectors driving the expression of a reporter gene from each of these two IRESs indicated that DAP5 preferentially promotes translation from the second IRES residing in the coding sequence. Furthermore, p53 mRNA expressed from a plasmid carrying this second IRES was selectively shifted to lighter polysomes upon DAP5 knockdown. Consequently, Delta 40p53 protein levels and the subsequent transcriptional activation of the 14-3-3 sigma gene, a known target of Delta 40p53, were strongly reduced. In addition, we show here that DAP5 interacts with p53 IRES elements in in vitro and in vivo binding studies, proving for the first time that DAP5 directly binds a target mRNA. Thus, through its ability to regulate IRES-dependent translation of the p53 mRNA, DAP5 may control the ratio between different p53 isoforms encoded by a single mRNA.

A framework for post-silicon realization of arbitrary instruction extensions on reconfigurable data-paths

In this paper we present a framework for realizing arbitrary instruction set extensions (IE) that are identified post-silicon. The proposed framework has two components viz., an IE synthesis methodology and the architecture of a reconfigurable data-path for realization of the such IEs. The IE synthesis methodology ensures maximal utilization of resources on the reconfigurable data-path. In this context we present the techniques used to realize IEs for applications that demand high throughput or those that must process data streams. The reconfigurable hardware called HyperCell comprises a reconfigurable execution fabric. The fabric is a collection of interconnected compute units. A typical use case of HyperCell is where it acts as a co-processor with a host and accelerates execution of IEs that are defined post-silicon. We demonstrate the effectiveness of our approach by evaluating the performance of some well-known integer kernels that are realized as IEs on HyperCell. Our methodology for realizing IEs through HyperCells permits overlapping of potentially all memory transactions with computations. We show significant improvement in performance for streaming applications over general purpose processor based solutions, by fully pipelining the data-path. (C) 2014 Elsevier B.V. All rights reserved.

Is the cellular initiation of translation an exclusive property of the initiator tRNAs?

Translation of mRNAs is the primary function of the ribosomal machinery. Although cells allow for a certain level of translational errors/mistranslation (which may well be a strategic need), maintenance of the fidelity of translation is vital for the cellular function and fitness. The P-site bound initiator tRNA selects the start codon in an mRNA and specifies the reading frame. A direct P-site binding of the initiator tRNA is a function of its special structural features, ribosomal elements, and the initiation factors. A highly conserved feature of the 3 consecutive G:C base pairs (3GC pairs) in the anticodon stem of the initiator tRNAs is vital in directing it to the P-site. Mutations in the 3GC pairs diminish/abolish initiation under normal physiological conditions. Using molecular genetics approaches, we have identified conditions that allow initiation with the mutant tRNAs in Escherichia coli. During our studies, we have uncovered a novel phenomenon of in vivo initiation by elongator tRNAs. Here, we recapitulate how the cellular abundance of the initiator tRNA, and nucleoside modifications in rRNA are connected with the tRNA selection in the P-site. We then discuss our recent finding of how a conserved feature in the mRNA, the Shine-Dalgarno sequence, influences tRNA selection in the P-site.

Synthesis of Instruction Extensions on HyperCell, a Reconfigurable Datapath

In this paper we present HyperCell as a reconfigurable datapath for Instruction Extensions (IEs). HyperCell comprises an array of compute units laid over a switch network. We present an IE synthesis methodology that enables post-silicon realization of IE datapaths on HyperCell. The synthesis methodology optimally exploits hardware resources in HyperCell to enable software pipelined execution of IEs. Exploitation of temporal reuse of data in HyperCell results in significant reduction of input/output bandwidth requirements of HyperCell.

Translation Induction on Indian Language Corpora Using Translingual Themes from Other Languages

Identifying translations from comparable corpora is a well-known problem with several applications, e.g. dictionary creation in resource-scarce languages. Scarcity of high quality corpora, especially in Indian languages, makes this problem hard, e.g. state-of-the-art techniques achieve a mean reciprocal rank (MRR) of 0.66 for English-Italian, and a mere 0.187 for Telugu-Kannada. There exist comparable corpora in many Indian languages with other ``auxiliary'' languages. We observe that translations have many topically related words in common in the auxiliary language. To model this, we define the notion of a translingual theme, a set of topically related words from auxiliary language corpora, and present a probabilistic framework for translation induction. Extensive experiments on 35 comparable corpora using English and French as auxiliary languages show that this approach can yield dramatic improvements in performance (e.g. MRR improves by 124% to 0.419 for Telugu-Kannada). A user study on WikiTSu, a system for cross-lingual Wikipedia title suggestion that uses our approach, shows a 20% improvement in the quality of titles suggested.

Discriminative language model adaptation for Mandarin broadcast speech transcription and translation

This paper investigates unsupervised test-time adaptation of language models (LM) using discriminative methods for a Mandarin broadcast speech transcription and translation task. A standard approach to adapt interpolated language models to is to optimize the component weights by minimizing the perplexity on supervision data. This is a widely made approximation for language modeling in automatic speech recognition (ASR) systems. For speech translation tasks, it is unclear whether a strong correlation still exists between perplexity and various forms of error cost functions in recognition and translation stages. The proposed minimum Bayes risk (MBR) based approach provides a flexible framework for unsupervised LM adaptation. It generalizes to a variety of forms of recognition and translation error metrics. LM adaptation is performed at the audio document level using either the character error rate (CER), or translation edit rate (TER) as the cost function. An efficient parameter estimation scheme using the extended Baum-Welch (EBW) algorithm is proposed. Experimental results on a state-of-the-art speech recognition and translation system are presented. The MBR adapted language models gave the best recognition and translation performance and reduced the TER score by up to 0.54% absolute. © 2007 IEEE.

Improving speech transcription for Mandarin-english translation

This paper describes the development of the CU-HTK Mandarin Speech-To-Text (STT) system and assesses its performance as part of a transcription-translation pipeline which converts broadcast Mandarin audio into English text. Recent improvements to the STT system are described and these give Character Error Rate (CER) gains of 14.3% absolute for a Broadcast Conversation (BC) task and 5.1% absolute for a Broadcast News (BN) task. The output of these STT systems is then post-processed, so that it consists of sentence-like segments, and translated into English text using a Statistical Machine Translation (SMT) system. The performance of the transcription-translation pipeline is evaluated using the Translation Edit Rate (TER) and BLEU metrics. It is shown that improving both the STT system and the post-STT segmentations can lower the TER scores by up to 5.3% absolute and increase the BLEU scores by up to 2.7% absolute. © 2007 IEEE.