Sanjay Ghemawat's research while affiliated with Google Inc. and other places
What is this page?
This page lists the scientific contributions of an author, who either does not have a ResearchGate profile, or has not yet added these contributions to their profile.
It was automatically created by ResearchGate to create a record of this author's body of work. We create such pages to advance our goal of creating and maintaining the most comprehensive scientific repository possible. In doing so, we process publicly available (personal) data relating to the author as a member of the scientific community.
If you're a ResearchGate member, you can follow this page to keep up with this author's work.
If you are this author, and you don't want us to display this page anymore, please let us know.
It was automatically created by ResearchGate to create a record of this author's body of work. We create such pages to advance our goal of creating and maintaining the most comprehensive scientific repository possible. In doing so, we process publicly available (personal) data relating to the author as a member of the scientific community.
If you're a ResearchGate member, you can follow this page to keep up with this author's work.
If you are this author, and you don't want us to display this page anymore, please let us know.
Publications (53)
Large language models have been shown to achieve remarkable performance across a variety of natural language tasks using few-shot learning, which drastically reduces the number of task-specific training examples needed to adapt the model to a particular application. To further our understanding of the impact of scale on few-shot learning, we traine...
We present the design of a new large scale orchestration layer for accelerators. Our system, Pathways, is explicitly designed to enable exploration of new systems and ML research ideas, while retaining state of the art performance for current models. Pathways uses a sharded dataflow graph of asynchronous operators that consume and produce futures,...
Many recent machine learning models rely on fine-grained dynamic control flow for training and inference. In particular, models based on recurrent neural networks and on reinforcement learning depend on recurrence relations, data-dependent conditional execution, and other features that call for dynamic control flow. These applications benefit from...
TensorFlow is a machine learning system that operates at large scale and in heterogeneous environments. TensorFlow uses dataflow graphs to represent computation, shared state, and the operations that mutate that state. It maps the nodes of a dataflow graph across many machines in a cluster, and within a machine across multiple computational devices...
TensorFlow is an interface for expressing machine learning algorithms, and an implementation for executing such algorithms. A computation expressed using TensorFlow can be executed with little or no change on a wide variety of heterogeneous systems, ranging from mobile devices such as phones and tablets up to large-scale distributed systems of hund...
Methods for organizing and retrieving data values in a persistent data structure are provided. Data values are grouped into data blocks and pointers are obtained for each data block. In addition, one or more summaries, related to a properties of the data block, are created and associated with the data block's pointer. The summaries allow for a more...
A system may present information regarding a document and provide an option for removing the document. The system may also receive selection of the option and remove the document when the option is selected. The system may aggregate information regarding documents that have been removed by a group of users and assign scores to a set of documents ba...
TensorFlow [1] is an interface for expressing machine learning algorithms, and an implementation for executing such algorithms. A computation expressed using TensorFlow can be executed with little or no change on a wide variety of heterogeneous systems, ranging from mobile devices such as phones and tablets up to large-scale distributed systems of...
A system includes: an engaging post identifier for identifying and retrieving engaging posts; an extended network post identifier for identifying extended posts from an extended network; a combining module for creating a combined list of added posts from the engaging post and the extended posts, the combining module generating one or more ranked po...
Techniques for crawling hyperlinked documents are provided. Hyperlinked documents to be crawled are grouped by host and the host to be crawled next is selected according to a stall time of the host. The stall time can indicate the earliest time that the host should be crawled and the stall times can be a predetermined amount of time, vary by host a...
Upon receipt of a document request, a client assistant examines its cache for the document. If not successful, a server searches for the requested document in its cache. If the server copy is still not fresh or not found, the server seeks the document from its host. If the host cannot provide the copy, the server seeks it from a document repository...
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, relating to language models stored for digital language processing. In one aspect, a method includes the actions of generating a language model, including: receiving a collection of n-grams from a corpus, each n-gram of the collection having a corresp...
A search engine for searching a corpus improves the relevancy of the results by classifying multiple terms in a search query as a single semantic unit. A semantic unit locator of the search engine generates a subset of documents that are generally relevant to the query based on the individual terms within the query. Combinations of search terms tha...
Methods and apparatus consistent with the invention allow a user to submit an ambiguous search query and to receive relevant search results. Queries can be expressed using character sets and/or languages that are different from the character set and/or language of at least some of the data that is to be searched. A translation between these charact...
A computer system comprising one or more processors and memory groups a set of documents into a plurality of clusters. Each cluster includes one or more documents of the set of documents and a respective cluster of documents of the plurality of clusters includes respective cluster data corresponding to a plurality of documents including a first doc...
Spanner is Google’s scalable, multiversion, globally distributed, and synchronously replicated database. It is the first system to distribute data at global scale and support externally-consistent distributed transactions. This article describes how Spanner is structured, its feature set, the rationale underlying various design decisions, and a nov...
Spanner is Google’s scalable, multiversion, globally distributed, and synchronously replicated database. It is the first system to distribute data at global scale and support externally-consistent distributed transactions. This article describes how Spanner is structured, its feature set, the rationale underlying various design decisions, and a nov...
Web quotes are gathered from web pages that link to a web page of interest. The web quote may include text from the paragraphs that contain the hypertext links to the page of interest as well as text from other portions of the linked web page, such as text from a nearby header. The obtained web quotes may be ranked based on quality or relevance and...
Provided is a method and system for indexing documents in a collection of linked documents. A link log, including one or more pairings of source documents and target documents is accessed. A sorted anchor map, containing one or more target document to source document pairings, is generated. The pairings in the sorted anchor map are ordered based on...
Spanner is Google's scalable, multi-version, globally-distributed, and synchronously-replicated database. It is the first system to distribute data at global scale and support externally-consistent distributed transactions. This paper describes how Spanner is structured, its feature set, the rationale underlying various design decisions, and a nove...
MapReduce (MR) has emerged as a flexible data processing tool for different applications. It has emerged as a programming model for processing and generating large data sets. Users specify a map function that processes a key or value pair to generate a set of intermediate key or value pairs and a reduce function that merges all intermediate values...
With the availability of large amounts of training data relevant to speech recognition scenarios, scalability becomes a very productive way to improve language model performance. We present a technique that represents a back-off n-gram language model using arrays of integer values and thus renders it amenable to effective block compression. We prop...
Bigtable is a distributed storage system for managing structured data that is designed to scale to a very large size: petabytes of data across thousands of commodity servers. Many projects at Google store data in Bigtable, including web indexing, Google Earth, and Google Finance. These applications place very different demands on Bigtable, both in...
MapReduce is a programming model and an associated implementation for processing and generating large datasets that is amenable to a broad variety of real-world tasks. Users specify the computation in terms of a map and a reduce function, and the underlying runtime system automatically parallelizes the computation across large-scale clusters of mac...
We have designed and implemented the Google File Sys- tem, a scalable distributed file system for large distributed data-intensive applications. It provides fault tolerance while running on inexpensive commodity hardware, and it delivers high aggregate performance to a large number of clients. While sharing many of the same goals as previous dis- t...
This paper describes the DIGITAL Continuous Profiling Infrastructure, a sampling-based profiling system designed to run continuously on production systems. The system supports multiprocessors, works on unmodified executables, and collects profiles for entire systems, including user programs, shared libraries, and the operating system kernel. Sample...
technique [1, 26, 27]. In this method, client calls are directed to a single primary server, which communicates This paper describes the design and implementation of the with other backup servers and waits for them to respond Harp file system. Harp is a replicated Unix file system before replying to the client. The system masks failures by accessib...
To provide high availability for services such as mail or bulletin boards, data must be replicated. One way to guarantee consistency of replicated data is to force service operations to occur in the same order at all sites, but this approach is expensive. In this paper, we propose lazy replication as a way to preserve consistency by exploiting the...
To provide high availability for services such as mail or bulletin boards, data must be replicated. One way to guarantee consistency of replicated data is to force service operations to occur in the same order at all sites, but this approach is expensive. In this paper, we propose lazy replication as a way to preserve consistency by exploiting the...
ProfileMe, which can provide significant insight into the behavior of programs running on complex microprocessors (especially out-of-order processors) [2]. ProfileMe requires only modest hardware modifications and can be used by our DCPI tools in a way that collects detailed profile information without substantial profiling overhead. The DCPI profi...
Replicating various components of a system is a common technique for providing highly available services in the presence of failures. A replication scheme is a mechanism for organizing these replicas so that as a group they provide a service that has the same semantics as the original unreplicated service. Viewstamped replication is a new replicati...
To provide high availability for services such as mail or bulletin boards, data must be replicated. One way to guarantee consistency of replicated data is to force service operations to occur in the same order at all sites, but this approach is expensive. For some applications a weaker causal operation order can preserve consistency while providing...
This article describes the Digital Continuous Profiling Infrastructure, a sampling-based profiling system designed to run continuously on production systems. The system supports multiprocessors, works on unmodified executables, and collects profiles for entire systems, including user programs, shared libraries, and the operating system kernel. Samp...
This paper describes the DIGITAL Continuous Profiling Infrastructure, a sampling-based profiling system designed to run continuously on production systems. The system supports multiprocessors, works on unmodified executables, and collects profiles for entire systems, including user programs, shared libraries, and the operating system kernel. Sample...
This paper proposes efficient disk management techniques for such databases.
Thor is an object-oriented database system designed for use in a heterogeneous distributed environment. It provides highly-reliable and highly-available persistent storage for objects, and supports safe sharing of these objects by applications written in different programming languages.Safe heterogeneous sharing of long-lived objects requires encap...
This document provides an introduction to using Thor, an object oriented database, and its language Theta. This document applies to Thor Release 0.1. Topics include installation, writing Theta code to implement objects, writing C++ application code, and debugging.
Object-oriented databases store many small objects on disks. Disks perform poorly when reading and writing individual small objects. This thesis presents a new storage management architecture that substantially improves disk performance of a distributed object-oriented database system. The storage architecture is built around a large modified objec...
This document describes a new programming language called Theta. Theta is a sequential, strongly-typed, object-oriented language. It provides separate mechanisms for type hierarchy, inheritance, and parametric polymorphism. It also provides separate mechanisms for specifications, which define the interfaces of new abstractions, and code that implem...
this paper I propose three disk management strategies for object-oriented databases. These strategies are based on earlier work on file-systems. They differ from this earlier work in their support for a large number of small objects and non-sequential access patterns.
Thor is a new object-oriented database system being developed at MIT. It allows applications written in different programming languages, and possibly running on heterogeneous machines and operating systems, to share objects conveniently. Our goal is to provide safe sharing of objects with higher-level semantics than is typical for today's file syst...
We propose several new techniques for resource management in a replicated object server. By coordinating cache and disk usage among the replicas, these techniques increase throughput and reduce fetch latency. Cache splitting speeds up fetches by avoiding redundant cache entries, effectively increasing the cache size. Coordinated writing coordinates...
An object-oriented database provides persistent storage for a
large number of objects. These objects may be very small, and the access
patterns are likely to be not as uniform as the mostly sequential reads
and writes seen in file-systems. For example, the 007 benchmark for
object-oriented databases specifies a number of traversals that follow
poin...
Introduction Harp is a replicated Unix file system accessible via the VFS interface. It provides highly available and reliable storage for files and guarantees that file operations are executed atomically in spite of concurrency and failures. Replication enables Harp to safely trade disk accesses for network communication and thus to provide good p...
Harp is a replicated Unix file system accessible via the VFS
interface. It provides highly available and reliable storage for files
and guarantees that file operations are executed atomically in spite of
concurrency and failures. Replication enables Harp to safely trade disk
accesses for network communication and thus to provide good performance
bo...
Citations
... Another Python package, Net2Brain 21 , has functionality for loading pre-existing DNNs, extracting their activations to stimuli, analyzing the extracted activations, and comparing them to brain activity measurements, but is similarly limited to a pre-defined library of models, and can only extract the outputs of PyTorch modules, not every tensor operation in a model. Another approach is the feature_extraction module provided in Torchvision 22 , which works on novel models (not just a predefined library), and can extract the results of non-module operations, but does not support models with dynamic control flow, where the computational graph of a network can vary across forward passes 23,24 . Examples of such models include recurrent neural networks that execute a varying number of loops until a given criterion is reached 25 , reinforcement learning models where the agent's behavior can vary stochastically or based on model inputs 26 , and graph convolutional networks where the computational graph can vary across model inputs 27 www.nature.com/scientificreports/ ...
... We focus on implementing performance comparison among two main types of NN structures: CNNs and RNNs, with their variations and combinations. Four basic layer types and their variations in TensorFlow [35] were chosen for investigation: 2D convolution layer (Conv2D), 1D convolution layer (Conv1D), LSTM layer, GRU layer, and their combinations. ...
![[object Object]](https://i1.rgstatic.net/ii/profile.image/278669114265607-1443451302469_Q64/Eugene-Brevdo-2.jpg)
... Gray countered that the current trend should prioritize the affordability of clusters of computers for data management and processing, rather than the acquisition of the largest and quickest individual computer. In contrast to their predecessors, which were confined to local and dedicated networks, interconnection technologies are crucial in today's Internet-connected clusters [5]. This is because these clusters must operate in parallel, irrespective of their distances, in order to manipulate the data sets required by the applications. ...
... TensorFlow-Lite [7] is an example of a TinyML-based algorithm that is optimized for deployment on embedded devices [9]. It includes a number of features that make it well-suited for implementing TinyML, such as support for on-device ML, quantization and pruning of models to reduce their size and improve performance, and a small footprint that allows it to run on devices with limited memory and storage. ...
![[object Object]](https://i1.rgstatic.net/ii/profile.image/272670101078036-1442021026588_Q64/Zhifeng-Chen-9.jpg)
... To process the respiratory sound, we used MATLAB 2021b. All deep learning models were developed using Python 3.8 with Tensorflow [28]. The models were trained on an NVIDIA RTX A5000 with 24 GB of GDDR6 RAM. ...
... The Google Earth Engine relies on multiple technologies within the Google data center, such as Flume Java framework, Bigtable, Spanner distributed databases [15], [16], Borg cluster management system, Colossus, and Google Fusion Tables [17], [18]. GEE provides accessibility via its API and web-based interface, enabling swift prototyping and visualization of outcomes. ...
... Sanjay Ghemawat presents a new storage management architecture that substantially improves disk performance of a distributed object-oriented database system [11]. The storage architecture is built around a large modified object buffer (MOB) that is stored in primary memory. ...
... The impact of profiling in non-deterministic systems: Most of the prediction models are forecast based on historical knowledge for short-term requests (Mallick et al., 2012). The historical knowledge token from the monitoring service, which logs information continuously as a profile in a searchable database (Anderson et al., 1997) and the whole process showed in Fig. 3. The proper analysis tool dissects the stored profile information at several levels. ...
![[object Object]](https://i1.rgstatic.net/ii/profile.image/277048154181634-1443064835211_Q64/William-Weihl.jpg)
... Google Dataset Search (GOODS) [67], [68] supports heterogeneous data storage used in Google's key-value stores Bigtable [22], file systems, and Spanner [28]; see also Sec. 4.4. ...
![[object Object]](https://i1.rgstatic.net/ii/profile.image/292467367792645-1446741062328_Q64/Dale-Woodford.jpg)
... We utilize Raft log [22] to enable transaction restarts after a crash by log replay. The protocol is similar to the existing one over Paxos [14] (e.g., MegaStore [1] and Spanner [4]) but we add external storage as transaction participants. Objcache precisely tracks updates of each chunk of all the objects at local storage and safely commits changes after confirming the completion of remote uploads. ...
