arXiv subjects

Hiren Patel

Publications and source records attributed to Hiren Patel.

15 recordsLinked to original sources

arXiv2025

ARM SVE Unleashed: Performance and Insights Across HPC Applications on Nvidia Grace

Vector architectures are essential for boosting computing throughput. ARM provides SVE as the next-generation length-agnostic vector extension beyond traditional fixed-length SIMD. This work provides a first study of the maturity and readiness of exploiting ARM and SVE in HPC. Using selected performance hardware events on the ARM Grace processor and analytical models, we derive new metrics to quantify the effectiveness of exploiting SVE vectorization to reduce executed instructions and improve performance speedup. We further propose an adapted roofline model that combines vector length and data elements to identify potential performance bottlenecks. Finally, we propose a decision tree for classifying the SVE-boosted performance in applications.

cs.DC↗

arXiv2024

From Diagnostic CT to DTI Tractography labels: Using Deep Learning for Corticospinal Tract Injury Assessment and Outcome Prediction in Intracerebral Haemorrhage

The preservation of the corticospinal tract (CST) is key to good motor recovery after stroke. The gold standard method of assessing the CST with imaging is diffusion tensor tractography. However, this is not available for most intracerebral haemorrhage (ICH) patients. Non-contrast CT scans are routinely available in most ICH diagnostic pipelines, but delineating white matter from a CT scan is challenging. We utilise nnU-Net, trained on paired diagnostic CT scans and high-directional diffusion tractography maps, to segment the CST from diagnostic CT scans alone, and we show our model reproduces diffusion based tractography maps of the CST with a Dice similarity coefficient of 57%. Surgical haematoma evacuation is sometimes performed after ICH, but published clinical trials to date show that whilst surgery reduces mortality, there is no evidence of improved functional recovery. Restricting surgery to patients with an intact CST may reveal a subset of patients for whom haematoma evacuation improves functional outcome. We investigated the clinical utility of our model in the MISTIE III clinical trial dataset. We found that our model's CST integrity measure significantly predicted outcome after ICH in the acute and chronic time frames, therefore providing a prognostic marker for patients to whom advanced diffusion tensor imaging is unavailable. This will allow for future probing of subgroups who may benefit from surgery.

eess.IV↗

arXiv2024

Towards Building Autonomous Data Services on Azure

Modern cloud has turned data services into easily accessible commodities. With just a few clicks, users are now able to access a catalog of data processing systems for a wide range of tasks. However, the cloud brings in both complexity and opportunity. While cloud users can quickly start an application by using various data services, it can be difficult to configure and optimize these services to gain the most value from them. For cloud providers, managing every aspect of an ever-increasing set of data services, while meeting customer SLAs and minimizing operational cost is becoming more challenging. Cloud technology enables the collection of significant amounts of workload traces and system telemetry. With the progress in data science (DS) and machine learning (ML), it is feasible and desirable to utilize a data-driven, ML-based approach to automate various aspects of data services, resulting in the creation of autonomous data services. This paper presents our perspectives and insights on creating autonomous data services on Azure. It also covers the future endeavors we plan to undertake and unresolved issues that still need attention.

cs.DC↗

arXiv2022

Deploying a Steered Query Optimizer in Production at Microsoft

Modern analytical workloads are highly heterogeneous and massively complex, making generic query optimizers untenable for many customers and scenarios. As a result, it is important to specialize these optimizers to instances of the workloads. In this paper, we continue a recent line of work in steering a query optimizer towards better plans for a given workload, and make major strides in pushing previous research ideas to production deployment. Along the way we solve several operational challenges including, making steering actions more manageable, keeping the costs of steering within budget, and avoiding unexpected performance regressions in production. Our resulting system, QQ-advisor, essentially externalizes the query planner to a massive offline pipeline for better exploration and specialization. We discuss various aspects of our design and show detailed results over production SCOPE workloads at Microsoft, where the system is currently enabled by default.

cs.DB↗

arXiv2022

Secure and Lightweight Strong PUF Challenge Obfuscation with Keyed Non-linear FSR

We propose a secure and lightweight key based challenge obfuscation for strong PUFs. Our architecture is designed to be resilient against learning attacks. Our obfuscation mechanism uses non-linear feedback shift registers (NLFSRs). Responses are directly provided to the user, without error correction or extra post-processing steps. We also discuss the cost of protecting our architecture against power analysis attacks with clock randomization, and Boolean masking. Security against learning attacks is assessed using avalanche criterion, and deep-neural network attacks. We designed a testchip in 65 nm CMOS. When compared to the baseline arbiter PUF implementation, the cost increase of our proposed architecture is 1.27x, and 2.2x when using clock randomization, and Boolean masking, respectively.

cs.CR↗

arXiv2022

Design Exploration and Security Assessment of PUF-on-PUF Implementations

We design, implement, and assess the security of several variations of the PUF-on-PUF (POP) architecture. We perform extensive experiments with deep neural networks (DNNs), showing results that endorse its resilience to learning attacks when using APUFs with 6, or more, stages in the first layer. Compositions using APUFs with 2, and 4 stages are shown vulnerable to DNN attacks. We reflect on such results, extending previous techniques of influential bits to assess stage bias in APUF instances. Our data shows that compositions not always preserve security properties of PUFs, the size of PUFs used plays a crucial role. We implemented a testchip in 65 nm CMOS to obtain accurate measurements of uniformity, uniqueness, and response stability for our POP implementations. Measurement results show that minimum bit error rate is obtained when using APUFs with 8 stages in the first layer, while fewer APUF stages lead to a large spread of bit error rate across different chips.

cs.CR↗

arXiv2022

Enhancing Strong PUF Security with Non-monotonic Response Quantization

Strong physical unclonable functions (PUFs) provide a low-cost authentication primitive for resource constrained devices. However, most strong PUF architectures can be modeled through learning algorithms with a limited number of CRPs. In this paper, we introduce the concept of non-monotonic response quantization for strong PUFs. Responses depend not only on which path is faster, but also on the distance between the arriving signals. Our experiments show that the resulting PUF has increased security against learning attacks. To demonstrate, we designed and implemented a non-monotonically quantized ring-oscillator based PUF in 65 nm technology. Measurement results show nearly ideal uniformity and uniqueness, with bit error rate of 13.4% over the temperature range from 0 C to 50 C.

eess.SY↗

arXiv2022

Predictable Sharing of Last-level Cache Partitions for Multi-core Safety-critical Systems

Last-level cache (LLC) partitioning is a technique to provide temporal isolation and low worst-case latency (WCL) bounds when cores access the shared LLC in multicore safety-critical systems. A typical approach to cache partitioning involves allocating a separate partition to a distinct core. A central criticism of this approach is its poor utilization of cache storage. Today's trend of integrating a larger number of cores exacerbates this issue such that we are forced to consider shared LLC partitions for effective deployments. This work presents an approach to share LLC partitions among multiple cores while being able to provide low WCL bounds.

cs.AR↗

arXiv2021

Phoebe: A Learning-based Checkpoint Optimizer

Easy-to-use programming interfaces paired with cloud-scale processing engines have enabled big data system users to author arbitrarily complex analytical jobs over massive volumes of data. However, as the complexity and scale of analytical jobs increase, they encounter a number of unforeseen problems, hotspots with large intermediate data on temporary storage, longer job recovery time after failures, and worse query optimizer estimates being examples of issues that we are facing at Microsoft. To address these issues, we propose Phoebe, an efficient learning-based checkpoint optimizer. Given a set of constraints and an objective function at compile-time, Phoebe is able to determine the decomposition of job plans, and the optimal set of checkpoints to preserve their outputs to durable global storage. Phoebe consists of three machine learning predictors and one optimization module. For each stage of a job, Phoebe makes accurate predictions for: (1) the execution time, (2) the output size, and (3) the start/end time taking into account the inter-stage dependencies. Using these predictions, we formulate checkpoint optimization as an integer programming problem and propose a scalable heuristic algorithm that meets the latency requirement of the production environment. We demonstrate the effectiveness of Phoebe in production workloads, and show that we can free the temporary storage on hotspots by more than 70% and restart failed jobs 68% faster on average with minimum performance impact. Phoebe also illustrates that adding multiple sets of checkpoints is not cost-efficient, which dramatically reduces the complexity of the optimization.

cs.DB↗

arXiv2021

Optimal Resource Allocation for Serverless Queries

Optimizing resource allocation for analytical workloads is vital for reducing costs of cloud-data services. At the same time, it is incredibly hard for users to allocate resources per query in serverless processing systems, and they frequently misallocate by orders of magnitude. Unfortunately, prior work focused on predicting peak allocation while ignoring aggressive trade-offs between resource allocation and run-time. Additionally, these methods fail to predict allocation for queries that have not been observed in the past. In this paper, we tackle both these problems. We introduce a system for optimal resource allocation that can predict performance with aggressive trade-offs, for both new and past observed queries. We introduce the notion of a performance characteristic curve (PCC) as a parameterized representation that can compactly capture the relationship between resources and performance. To tackle training data sparsity, we introduce a novel data augmentation technique to efficiently synthesize the entire PCC using a single run of the query. Lastly, we demonstrate the advantages of a constrained loss function coupled with GNNs, over traditional ML methods, for capturing the domain specific behavior through an extensive experimental evaluation over SCOPE big data workloads at Microsoft.

cs.DB↗

arXiv2020

Cost Models for Big Data Query Processing: Learning, Retrofitting, and Our Findings

Query processing over big data is ubiquitous in modern clouds, where the system takes care of picking both the physical query execution plans and the resources needed to run those plans, using a cost-based query optimizer. A good cost model, therefore, is akin to better resource efficiency and lower operational costs. Unfortunately, the production workloads at Microsoft show that costs are very complex to model for big data systems. In this work, we investigate two key questions: (i) can we learn accurate cost models for big data systems, and (ii) can we integrate the learned models within the query optimizer. To answer these, we make three core contributions. First, we exploit workload patterns to learn a large number of individual cost models and combine them to achieve high accuracy and coverage over a long period. Second, we propose extensions to Cascades framework to pick optimal resources, i.e, number of containers, during query planning. And third, we integrate the learned cost models within the Cascade-style query optimizer of SCOPE at Microsoft. We evaluate the resulting system, Cleo, in a production environment using both production and TPC-H workloads. Our results show that the learned cost models are 2 to 3 orders of magnitude more accurate, and 20X more correlated with the actual runtimes, with a large majority (70%) of the plan changes leading to substantial improvements in latency as well as resource usage.

cs.DB↗

arXiv2019

Cloudy with high chance of DBMS: A 10-year prediction for Enterprise-Grade ML

Machine learning (ML) has proven itself in high-value web applications such as search ranking and is emerging as a powerful tool in a much broader range of enterprise scenarios including voice recognition and conversational understanding for customer support, autotuning for videoconferencing, intelligent feedback loops in large-scale sysops, manufacturing and autonomous vehicle management, complex financial predictions, just to name a few. Meanwhile, as the value of data is increasingly recognized and monetized, concerns about securing valuable data and risks to individual privacy have been growing. Consequently, rigorous data management has emerged as a key requirement in enterprise settings. How will these trends (ML growing popularity, and stricter data governance) intersect? What are the unmet requirements for applying ML in enterprise settings? What are the technical challenges for the DB community to solve? In this paper, we present our vision of how ML and database systems are likely to come together, and early steps we take towards making this vision a reality.

cs.DB↗

arXiv2018

HourGlass: Predictable Time-based Cache Coherence Protocol for Dual-Critical Multi-Core Systems

We present a hardware mechanism called HourGlass to predictably share data in a multi-core system where cores are explicitly designated as critical or non-critical. HourGlass is a time-based cache coherence protocol for dual-critical multi-core systems that ensures worst-case latency (WCL) bounds for memory requests originating from critical cores. Although HourGlass does not provide either WCL or bandwidth guarantees for memory requests from non-critical cores, it promotes the use of timers to improve its bandwidth utilization while still maintaining WCL bounds for critical cores. This encourages a trade-off between the WCL bounds for critical cores, and the improved memory bandwidth for non-critical cores via timer configurations. We evaluate HourGlass using gem5, and with multithreaded benchmark suites including SPLASH-2, and synthetic workloads. Our results show that the WCL for critical cores with HourGlass is always within the analytical WCL bounds, and provides a tighter WCL bound on critical cores compared to the state-of-the-art real-time cache coherence protocol. Further, we show that HourGlass enables a trade-off between provable WCL bounds for critical cores, and improved bandwidth utilization for non-critical cores. The average-case performance of HourGlass is comparable to the state-of-the-art real-time cache coherence protocol, and suffers a slowdown of 1.43x and 1.46x compared to the conventional MSI and MESI protocols.

cs.AR↗

arXiv2017

Transplanckian Censorship and Global Cosmic Strings

Large field excursions are required in a number of axion models of inflation. These models also possess global cosmic strings, around which the axion follows a path mirroring the inflationary trajectory. Cosmic strings are thus an interesting theoretical laboratory for the study of transplanckian field excursions. We describe connections between various effective field theory models of axion monodromy and study the classical spacetimes around their supercritical cosmic strings. For small decay constants $f M_p/f$, the EFT is under control and the string cores undergo topological inflation, which may be either of exponential or power-law type. We show that the exterior spacetime is nonsingular and equivalent to a decompactifying cigar geometry, with the radion rolling in a potential generated by axion flux. Signals are able to circumnavigate infinite straight strings in finite but exponentially long time, $t\sim e^{Δa/M_p}$. For finite loops of supercritical string in asymptotically flat space, we argue that if topological inflation occurs, then topological censorship implies transplanckian censorship, or that external observers are forbidden from threading the loop and observing the full excursion of the axion.

hep-th↗

arXiv2010

Vacuum Stability, Perturbativity, and Scalar Singlet Dark Matter

We analyze the one-loop vacuum stability and perturbativity bounds on a singlet extension of the Standard Model (SM) scalar sector containing a scalar dark matter candidate. We show that the presence of the singlet-doublet quartic interaction relaxes the vacuum stability lower bound on the SM Higgs mass as a function of the cutoff and lowers the corresponding upper bound based on perturbativity considerations. We also find that vacuum stability requirements may place a lower bound on the singlet dark matter mass for given singlet quartic self coupling, leading to restrictions on the parameter space consistent with the observed relic density. We argue that discovery of a light singlet scalar dark matter particle could provide indirect information on the singlet quartic self-coupling.

hep-ph↗