This document proposes a read-write split near-line data loading method and architecture to:
- Increase data loading performance by separating write operations from read operations. A WriteServer handles write requests and loads data to HDFS to be read from by RegionServers.
- Control resources used by write operations to ensure read operations are not starved of resources like CPU, network, disk I/O, and handlers.
- Provide an architecture corresponding to Kafka and HDFS for streaming data from Kafka to HDFS to be loaded into HBase in a delayed manner.
- Include optimizations like task balancing across WriteServer slaves, prioritized compaction of small files, and customizable storage engines.
- Report test results showing one Write
Apache HBase, Accelerated: In-Memory Flush and Compaction HBaseCon
Eshcar Hillel and Anastasia Braginsky (Yahoo!)
Real-time HBase application performance depends critically on the amount of I/O in the datapath. Here we’ll describe an optimization of HBase for high-churn applications that frequently insert/update/delete the same keys, such as for high-speed queuing and e-commerce.
HBaseCon 2012 | Base Metrics: What They Mean to You - ClouderaCloudera, Inc.
If you’re running an HBase cluster in production, you’ve probably noticed that HBase shares a number of useful metrics about everything from your block cache performance to your HDFS latencies over JMX (or Ganglia, or just a file). The problem is that it’s sometimes hard to know what these metrics mean to you and your users. Should you be worried if your memstore SizeMB is 1.5GB? What if your regionservers have a hundred stores each? This talk will explain how to understand and interpret the metrics HBase exports. Along the way we’ll cover some high-level background on HBase’s internals, and share some battle tested rules-of-thumb about how to interpret and react to metrics you might see.
Speaker: Bryan Beaudreault (HubSpot)
Running HBase in real time in the cloud provides an interesting and ever-changing set of challenges -- instance types are not ideal, neighbors can degrade your performance, and instances can randomly die in unanticipated ways. This talk will cover what HubSpot has learned about running in production on Amazon EC2, how it handle DR and redundancy, and the tooling the team has found to be the most helpful.
HBaseCon 2015: HBase at Scale in an Online and High-Demand EnvironmentHBaseCon
Pinterest runs 38 different HBase clusters in production, doing a lot of different types of work—with some doing up to 5 million operations per second. In this talk, you'll get details about how we do capacity planning, maintenance tasks such as online automated rolling compaction, configuration management, and monitoring.
Speakers: Liang Xie and Honghua Feng (Xiamoi)
This talk covers the HBase environment at Xiaomi, including thoughts and practices around latency, hardware/OS/VM configuration, GC tuning, the use of a new write thread model and reverse scan, and block index optimization. It will also include some discussion of planned JIRAs based on these approaches.
HBase 2.0 is the next stable major release for Apache HBase scheduled for early 2017. It is the biggest and most exciting milestone release from the Apache community after 1.0. HBase-2.0 contains a large number of features that is long time in the development, some of which include rewritten region assignment, perf improvements (RPC, rewritten write pipeline, etc), async clients, C++ client, offheaping memstore and other buffers, Spark integration, shading of dependencies as well as a lot of other fixes and stability improvements. We will go into technical details on some of the most important improvements in the release, as well as what are the implications for the users in terms of API and upgrade paths. Existing users of HBase/Phoenix as well as operators managing HBase clusters will benefit the most where they can learn about the new release and the long list of features. We will also briefly cover earlier 1.x release lines and compatibility and upgrade paths for existing users and conclude by giving an outlook on the next level of initiatives for the project.
Breaking the Sound Barrier with Persistent Memory HBaseCon
Liqi Yi and Shylaja Kokoori (Intel)
A fully optimized HBase cluster could easily hit the limit of the underlying storage device’s capability, which is beyond the reach of software optimization alone. To get around this constraint, we need a new design that brings data processing and data storage closer together. In this presentation, we will look at how persistent memory will change the way large datasets are stored. We will review the hardware characteristics of 3D XPoint™, a new persistent memory technology with low latency and high capacity. We will also discuss opportunities for further improvement within the HBase framework using persistent memory.
In this session, you will learn the work Xiaomi has done to improve the availability and stability of our HBase clusters, including cross-site data and service backup and a coordinated compaction framework. You'll also learn about the Themis framework, which supports cross-row transactions on HBase based on Google's percolator algorithm, and its usage in Xiaomi's applications.
Countdown to PostgreSQL v9.5 - Foriegn Tables can be part of Inheritance Tree Ashnikbiz
Distributed databases and horizontal scale up is one of the key demands in today's date. PostgreSQL already had some vertical scaling features and horizontal scale-up by adding disks and table partitioning/child tables. With release of v9.5, PostgreSQL will get basic foundation for native sharing capability. From v9.5 Foreign Tables will be able to participate in Inheritance Tree as a child or parent table i.e. one can have table partitions residing on different system.
In our countdown to v9.5 series of hangouts, we will be covering some of the great features of PostgreSQL v9.5 and what is their real life applicability. In the first hangout in this series we will be talking about-
- The feature of foreign partitions/child tables
- Syntax and usage
- EXPLAIN plan demo
- Use cases and benefits
Join us for more and send us your queries on success@ashnik.com
The document describes Accordion, a novel in-memory compaction algorithm for HBase that improves write performance. Accordion applies the log-structured merge tree design used in LSM databases to HBase's memory data structure. This transforms random writes to sequential writes, keeping more data in memory for longer. As a result, Accordion achieves higher write throughput, lower disk I/O, and reduced read latency compared to HBase's previous approach. Accordion has been integrated into HBase 2.0 as the default memory compaction implementation.
HBaseCon 2013: Streaming Data into Apache HBase using Apache Flume: Experienc...Cloudera, Inc.
This document discusses using Apache Flume to stream data into Apache HBase. It describes how Flume provides a scalable and flexible way to collect and transport log and event data to HBase. Specifically, it covers the HBase sink plugin for Flume, which allows routing Flume events to HBase tables. It notes that while the initial HBase sink had limitations, the asynchronous HBase sink improved performance by fully utilizing the HBase cluster. Overall, the document presents Flume as a viable alternative to directly writing to HBase and provides flexibility to change schemas without code changes.
HBaseCon 2012 | Solbase - Kyungseog Oh, PhotobucketCloudera, Inc.
Solbase is an exciting new open-source, real-time search engine being developed at Photobucket to service the over 30 million daily search requests Photobucket handles. Solbase replaces Lucene’s file system-based index with HBase. This allows the system to update in real-time and linearly scale to serve millions of daily search requests on a large dataset. This session will explore the architecture of Solbase as well as some of Lucene/Solr’s inherent issues we overcame. Finally, we’ll go over performance metrics of Solbase against production traffic.
HBase Accelerated introduces an in-memory flush and compaction pipeline for HBase to improve performance of real-time workloads. By keeping data in memory longer and avoiding frequent disk flushes and compactions, it reduces I/O and improves read and scan latencies. Evaluation on workloads with high update rates and small working sets showed the new approach significantly outperformed the default HBase implementation by serving most data from memory. Work is ongoing to further optimize the in-memory representation and memory usage.
HBase and HDFS: Understanding FileSystem Usage in HBaseenissoz
This document discusses file system usage in HBase. It provides an overview of the three main file types in HBase: write-ahead logs (WALs), data files, and reference files. It describes durability semantics, IO fencing techniques for region server recovery, and how HBase leverages data locality through short circuit reads, checksums, and block placement hints. The document is intended help understand HBase's interactions with HDFS for tuning IO performance.
HBaseCon 2012 | Learning HBase Internals - Lars Hofhansl, SalesforceCloudera, Inc.
The strength of an open source project resides entirely in its developer community; a strong democratic culture of participation and hacking makes for a better piece of software. The key requirement is having developers who are not only willing to contribute, but also knowledgeable about the project’s internal structure and architecture. This session will introduce developers to the core internal architectural concepts of HBase, not just “what” it does from the outside, but “how” it works internally, and “why” it does things a certain way. We’ll walk through key sections of code and discuss key concepts like the MVCC implementation and memstore organization. The goal is to convert serious “HBase Users” into HBase Developer Users”, and give voice to some of the deep knowledge locked in the committers’ heads.
Speaker: Vladimir Rodionov (bigbase.org)
This talks introduces a totally new implementation of a multilayer caching in HBase called BigBase. BigBase has a big advantage over HBase 0.94/0.96 because of an ability to utilize all available server RAM in the most efficient way, and because of a novel implementation of a L3 level cache on fast SSDs. The talk will show that different type of caches in BigBase work best for different type of workloads, and that a combination of these caches (L1/L2/L3) increases the overall performance of HBase by a very wide margin.
Various HA and DR setups for Postgres Plus Advanced Server -
Active – Passive OS HA Clustering
Log Shipping Replication (Hot Standby Mode)
Hot Streaming Replication (Hot Standby Mode)
EDB Postgres Plus Failover Manager
HA with read scaling (with pg-pool)
xDB Single Master Replication (SMR)
xDB Multi Master Replication (MMR)
Use Cases
HBase release managers Lars Hofhansl, Andrew Purtell, Enis Soztutar, Michael Stack, and Liyin Tang jointly present highlights from their releases, and take your questions throughout.
HBaseCon 2015: Taming GC Pauses for Large Java Heap in HBaseHBaseCon
In this presentation, we will introduce Hotspot's Garbage First collector (G1GC) as the most suitable collector for latency-sensitive applications running with large memory environments. We will first discuss G1GC internal operations and tuning opportunities, and also cover tuning flags that set desired GC pause targets, change adaptive GC thresholds, and adjust GC activities at runtime. We will provide several HBase case studies using Java heaps as large as 100GB that show how to best tune applications to remove unpredicted, protracted GC pauses.
HDFS allows storing large amounts of data across multiple machines by splitting files into blocks and replicating those blocks for reliability. It addresses challenges of big data like volume, velocity, and variety by providing a distributed storage solution that scales horizontally. Traditional systems are limited by network bandwidth, storage capacity of individual machines, and single points of failure. HDFS introduces a scalable architecture with a master NameNode and slave DataNodes that stores data blocks, addressing these issues through data distribution and fault tolerance.
Still All on One Server: Perforce at Scale Perforce
Google runs the busiest single Perforce server on the planet, and one of the largest repositories in any source control system. This session will address server performance and other issues of scale, as well as where Google is in general, how it got there and how it continues to stay ahead of its users.
With Hadoop-3.0.0-alpha2 being released in January 2017, it's time to have a closer look at the features and fixes of Hadoop 3.0.
We will have a look at Core Hadoop, HDFS and YARN, and answer the emerging question whether Hadoop 3.0 will be an architectural revolution like Hadoop 2 was with YARN & Co. or will it be more of an evolution adapting to new use cases like IoT, Machine Learning and Deep Learning (TensorFlow)?
Big Data and Hadoop in Cloud - Leveraging Amazon EMRVijay Rayapati
This document discusses big data, Hadoop, and using Hadoop in the cloud via Amazon EMR. It provides an overview of big data and what Hadoop is, explains how Hadoop works and how it can help store and process large datasets. It then discusses how Amazon EMR can be used to deploy Hadoop clusters in the cloud without having to manage the underlying infrastructure, and provides instructions on setting up and using EMR. Finally, it discusses debugging, profiling, and performance tuning Hadoop jobs and EMR clusters.
Ibm spectrum scale fundamentals workshop for americas part 4 Replication, Str...xKinAnx
The document provides an overview of IBM Spectrum Scale Active File Management (AFM). AFM allows data to be accessed globally across multiple clusters as if it were local by automatically managing asynchronous replication. It describes the various AFM modes including read-only caching, single-writer, and independent writer. It also covers topics like pre-fetching data, cache eviction, cache states, expiration of stale data, and the types of data transferred between home and cache sites.
The document summarizes a presentation on optimizing Linux, Windows, and Firebird for heavy workloads. It describes two customer implementations using Firebird - a medical company with 17 departments and over 700 daily users, and a repair services company with over 500 daily users. It discusses tuning the operating system, hardware, CPU, RAM, I/O, network, and Firebird configuration to improve performance under heavy loads. Specific recommendations are provided for Linux and Windows configuration.
Hadoop is an open-source framework that processes large datasets in a distributed manner across commodity hardware. It uses a distributed file system (HDFS) and MapReduce programming model to store and process data. Hadoop is highly scalable, fault-tolerant, and reliable. It can handle data volumes and variety including structured, semi-structured and unstructured data.
Cloud computing UNIT 2.1 presentation inRahulBhole12
Cloud storage allows users to store files online through cloud storage providers like Apple iCloud, Dropbox, Google Drive, Amazon Cloud Drive, and Microsoft SkyDrive. These providers offer various amounts of free storage and options to purchase additional storage. They allow files to be securely uploaded, accessed, and synced across devices. The best cloud storage provider depends on individual needs and preferences regarding storage space requirements and features offered.
Updated version of my talk about Hadoop 3.0 with the newest community updates.
Talk given at the codecentric Meetup Berlin on 31.08.2017 and on Data2Day Meetup on 28.09.2017 in Heidelberg.
- The document discusses running Hive/Spark on S3 object storage using S3A committers and running HBase on NFS file storage instead of HDFS. This separates compute and storage and avoids HDFS operations and complexity. S3A committers allow fast, atomic writes to S3 without renaming files. Benchmark results show the magic committer is faster than the file committer for S3 writes. HBase performance tests show FlashBlade NFS providing low latency for random reads/writes compared to Amazon EFS.
This document summarizes Marian Marinov's testing and experience with different distributed filesystems at his company SiteGround. He tested CephFS, GlusterFS, MooseFS, OrangeFS, and BeeGFS. CephFS required a lot of resources but lacked redundancy. GlusterFS was relatively easy to set up but had high CPU usage. MooseFS and OrangeFS were also easy to set up. Ultimately, they settled on Ceph RBD with NFS and caching for performance and simplicity. File creation performance tests showed MooseFS and NFS+Ceph RBD outperformed OrangeFS and GlusterFS. Tuning settings like MTU, congestion control, and caching helped optimize performance.
Red Hat Storage Server Administration Deep DiveRed_Hat_Storage
"In this session for administrators of all skill levels, you’ll get a deep technical dive into Red Hat Storage Server and GlusterFS administration.
We’ll start with the basics of what scale-out storage is, and learn about the unique implementation of Red Hat Storage Server and its advantages over legacy and competing technologies. From the basic knowledge and design principles, we’ll move to a live start-to-finish demonstration. Your experience will include:
Building a cluster.
Allocating resources.
Creating and modifying volumes of different types.
Accessing data via multiple client protocols.
A resiliency demonstration.
Expanding and contracting volumes.
Implementing directory quotas.
Recovering from and preventing split-brain.
Asynchronous parallel geo-replication.
Behind-the-curtain views of configuration files and logs.
Extended attributes used by GlusterFS.
Performance tuning basics.
New and upcoming feature demonstrations.
Those new to the scale-out product will leave this session with the knowledge and confidence to set up their first Red Hat Storage Server environment. Experienced administrators will sharpen their skills and gain insights into the newest features. IT executives and managers will gain a valuable overview to help fuel the drive for next-generation infrastructures."
This document summarizes Marian Marinov's testing and experience with various distributed filesystems including CephFS, GlusterFS, MooseFS, OrangeFS, and BeeGFS. Some key findings are:
- CephFS requires significant resources but lacks redundancy for small clusters. GlusterFS offers redundancy but can have high CPU usage.
- MooseFS and OrangeFS were easy to setup but MooseFS offered better reliability and stats.
- Performance testing found MooseFS and NFS+Ceph to have better small file creation times than GlusterFS and OrangeFS. Network latency was identified as a major factor impacting distributed filesystem performance.
- Tuning efforts focused on NFS
The document discusses new features in Apache Hadoop Common and HDFS for version 3.0. Key updates include upgrading the minimum Java version to Java 8, improving dependency management, adding a new Azure Data Lake Storage connector, and introducing erasure coding in HDFS to improve storage efficiency. Erasure coding in HDFS phase 1 allows for striping of small blocks and parallel writes/reads while trading off higher network usage compared to replication.
In this session, we'll discuss architectural, design and tuning best practices for building rock solid and scalable Alfresco Solutions. We'll cover the typical use cases for highly scalable Alfresco solutions, like massive injection and high concurrency, also introducing 3.3 and 3.4 Transfer / Replication services for building complex high availability enterprise architectures.
Hadoop is an open-source framework for distributed storage and processing of large datasets across clusters of commodity hardware. It has several core components including HDFS for distributed file storage and MapReduce for distributed processing. HDFS stores data across clusters of machines with replication for fault tolerance. MapReduce allows parallel processing of large datasets in a distributed manner. Hadoop was designed with goals of using commodity hardware, easy recovery from failures, large distributed file systems, and fast processing of large datasets.
Red Hat Ceph Storage Acceleration Utilizing Flash Technology Red_Hat_Storage
Red Hat Ceph Storage can utilize flash technology to accelerate applications in three ways: 1) use all flash storage for highest performance, 2) use a hybrid configuration with performance critical data on flash tier and colder data on HDD tier, or 3) utilize host caching of critical data on flash. Benchmark results showed that using NVMe SSDs in Ceph provided much higher performance than SATA SSDs, with speed increases of up to 8x for some workloads. However, testing also showed that Ceph may not be well-suited for OLTP MySQL workloads due to small random reads/writes, as local SSD storage outperformed the Ceph cluster. Proper Linux tuning is also needed to maximize SSD performance within
Hadoop 3.0 will include major new features like HDFS erasure coding for improved storage efficiency and YARN support for long running services and Docker containers to improve resource utilization. However, it will maintain backwards compatibility and a focus on testing given the importance of compatibility for existing Hadoop users. The release is targeted for late 2017 after several alpha and beta stages.
Apache Hadoop 3 is coming! As the next major milestone for hadoop and big data, it attracts everyone's attention as showcase several bleeding-edge technologies and significant features across all components of Apache Hadoop: Erasure Coding in HDFS, Docker container support, Apache Slider integration and Native service support, Application Timeline Service version 2, Hadoop library updates and client-side class path isolation, etc. In this talk, first we will update the status of Hadoop 3.0 releasing work in apache community and the feasible path through alpha, beta towards GA. Then we will go deep diving on each new feature, include: development progress and maturity status in Hadoop 3. Last but not the least, as a new major release, Hadoop 3.0 will contain some incompatible API or CLI changes which could be challengeable for downstream projects and existing Hadoop users for upgrade - we will go through these major changes and explore its impact to other projects and users.
This document summarizes new file system and storage features in Red Hat Enterprise Linux (RHEL) 6 and 7. It discusses enhancements to logical volume management (LVM) such as thin provisioning and snapshots. It also covers expanded file system options like XFS, improvements to NFS including parallel NFS, and general performance enhancements.
Similar to hbaseconasia2017: Large scale data near-line loading method and architecture (20)
hbaseconasia2017: Building online HBase cluster of Zhihu based on KubernetesHBaseCon
Zhiyong Bai
As a high performance and scalable key value database, Zhihu use HBase to provide online data store system along with Mysql and Redis. Zhihu’s platform team had accumulated some experience in technology of container, and this time, based on Kubernetes, we build flexible platform of online HBase system, create multiple logic isolated HBase clusters on the shared physical cluster with fast rapid,and provide customized service for different business needs. Combined with Consul and DNS server, we implement high available access of HBase using client mainly written with Python. This presentation is mainly shared the architecture of online HBase platform in Zhihu and some practical experience in production environment.
hbaseconasia2017 hbasecon hbase
Jingcheng Du
Apache Beam is an open source and unified programming model for defining batch and streaming jobs that run on many execution engines, HBase on Beam is a connector that allows Beam to use HBase as a bounded data source and target data store for both batch and streaming data sets. With this connector HBase can work with many batch and streaming engines directly, for example Spark, Flink, Google Cloud Dataflow, etc. In this session, I will introduce Apache Beam, and the current implementation of HBase on Beam and the future plan on this.
hbaseconasia2017 hbasecon hbase
https://www.eventbrite.com/e/hbasecon-asia-2017-tickets-34935546159#
hbaseconasia2017: Removable singularity: a story of HBase upgrade in PinterestHBaseCon
Tianying Chang
HBase is used to serve online facing traffic in Pinterest. It means no downtime is allowed. However, we were on HBase 94. To upgrade to latest version, we need to figure out a way to live upgrade while keeping Pinterest site live. Recently, we successfully upgrade 94 HBase cluster to 1.2 with no downtime. We made change to both Asynchbase and HBase server side. We will talk about what we did and how we did it. We will also talk about the finding in config and performance tuning we did to achieve low latency.
hbaseconasia2017 hbasecon hbase https://www.eventbrite.com/e/hbasecon-asia-2017-tickets-34935546159#
hbaseconasia2017: Ecosystems with HBase and CloudTable service at HuaweiHBaseCon
CTBase is a lightweight HBase client designed for structured data use cases. It provides features like schematized tables, global secondary indexes, cluster tables for joins, and online schema changes. Tagram is a distributed bitmap index implementation on HBase that supports ad-hoc queries on low-cardinality attributes with millisecond latency. CloudTable Service offers HBase as a managed service on Huawei Cloud with features including easy maintenance, security, high performance, service level agreements, high availability and low cost.
hbaseconasia2017: HBase Practice At XiaoMiHBaseCon
Zheng Hu
We'll share some HBase experience at XiaoMi:
1. How did we tuning G1GC for HBase Clusters.
2. Development and performance of Async HBase Client.
hbaseconasia2017 hbasecon hbase xiaomi https://www.eventbrite.com/e/hbasecon-asia-2017-tickets-34935546159#
As HBase and Hadoop continue to become routine across enterprises, these enterprises inevitably shift priorities from effective deployments to cost-efficient operations. Consolidation of infrastructure, the sum of hardware, software, and system-administrator effort, is the most common strategy to reduce costs. As a company grows, the number of business organizations, development teams, and individuals accessing HBase grows commensurately, creating a not-so-simple requirement: HBase must effectively service many users, each with a variety of use-cases. This is problem is known as multi-tenancy. While multi-tenancy isn’t a new problem, it also isn’t a solved one, in HBase or otherwise. This talk will present a high-level view of the common issues organizations face when multiple users and teams share a single HBase instance and how certain HBase features were designed specifically to mitigate the issues created by the sharing of finite resources.
HBaseCon2017 Quanta: Quora's hierarchical counting system on HBaseHBaseCon
Hundreds of millions of people use Quora to find accurate, informative, and trustworthy answers to their questions. As it so happens, counting things at scale is both an important and a difficult problem to solve.
In this talk, we will be talking about Quanta, Quora's counting system built on top of HBase that powers our high-volume near-realtime analytics that serves many applications like ads, content views, and many dashboards. In addition to regular counting, Quanta supports count propagation along the edges of an arbitrary DAG. HBase is the underlying data store for both the counting data and the graph data.
We will describe the high-level architecture of Quanta and share our design goals, constraints, and choices that enabled us to build Quanta very quickly on top of our existing infrastructure systems.
In the age of NoSQL, big data storage engines such as HBase have given up ACID semantics of traditional relational databases, in exchange for high scalability and availability. However, it turns out that in practice, many applications require consistency guarantees to protect data from concurrent modification in a massively parallel environment. In the past few years, several transaction engines have been proposed as add-ons to HBase; three different engines, namely Omid, Tephra, and Trafodion were open-sourced in Apache alone. In this talk, we will introduce and compare the different approaches from various perspectives including scalability, efficiency, operability and portability, and make recommendations pertaining to different use cases.
In order to effectively predict and prevent online fraud in real time, Sift Science stores hundreds of terabytes of data in HBase—and needs it to be always available. This talk will cover how we used circuit-breaking, cluster failover, monitoring, and automated recovery procedures to improve our HBase uptime from 99.7% to 99.99% on top of unreliable cloud hardware and networks.
In DiDi Chuxing Company, which is China’s most popular ride-sharing company. we use HBase to serve when we have a bigdata problem.
We run three clusters which serve different business needs. We backported the Region Grouping feature back to our internal HBase version so we could isolate the different use cases.
We built the Didi HBase Service platform which is popular amongst engineers at our company. It includes a workflow and project management function as well as a user monitoring view.
Internally we recommend users use Phoenix to simplify access.even more,we used row timestamp;multidimensional table schema to slove muti dimension query problems
C++, Go, Python, and PHP clients get to HBase via thrift2 proxies and QueryServer.
We run many important buisness applications out of our HBase cluster such as ETA/GPS/History Order/API metrics monitoring/ and Traffic in the Cloud. If you are interested in any aspects listed above, please come to our talk. We would like to share our experiences with you.
HBaseCon2017 gohbase: Pure Go HBase ClientHBaseCon
gohbase is an implementation of an HBase client in pure Go: https://github.com/tsuna/gohbase. In this presentation we'll talk about its architecture and compare its performance against the native Java HBase client as well as AsyncHBase (http://opentsdb.github.io/asynchbase/) and some nice characteristics of golang that resulted in a simpler implementation.
HBaseCon2017 Improving HBase availability in a multi tenant environmentHBaseCon
The document discusses improvements made by Hubspot's Big Data Team to increase the availability of HBase in a multi-tenant environment. It outlines reducing the cost of region server failures by improving mean time to recovery, addressing issues that slowed recovery, and optimizing the load balancer. It also details eliminating workload-driven failures through service limits and improving hardware monitoring to reduce impacts of failures. The changes resulted in 8-10x faster balancing, reduced recovery times from 90 to 30 seconds, and consistently achieving 99.99% availability across clusters.
HBaseCon2017 Spark HBase Connector: Feature Rich and Efficient Access to HBas...HBaseCon
Both Spark and HBase are widely used, but how to use them together with high performance and simplicity is a very hard topic. Spark HBase Connector(SHC) provides feature rich and efficient access to HBase through Spark SQL. It bridges the gap between the simple HBase key value store and complex relational SQL queries and enables users to perform complex data analytics on top of HBase using Spark.
SHC implements the standard Spark data source APIs, and leverages the Spark catalyst engine for query optimization. To achieve high performance, SHC constructs the RDD from scratch instead of using the standard HadoopRDD. With the customized RDD, all critical techniques can be applied and fully implemented, such as partition pruning, column pruning, predicate pushdown and data locality. The design makes the maintenance very easy, while achieving a good tradeoff between performance and simplicity.
Also, SHC has supported Phoenix data as input to HBase in addition to Avro data. Defaulting to a simple native binary encoding seems susceptible to future changes and is a risk for users who write data from SHC into HBase. For example, with SHC going forward, backwards compatibility needs to be properly handled. So the default, SHC needs to support a more standard and well tested format like Phoenix.
In this talk, we will demo how SHC works, how to use SHC in secure/non-secure clusters, how SHC works with multi-HBase clusters, etc. This talk will also benefit people who use Spark and other data sources (besides HBase) as it inspires them with ideas of how to support high performance data source access at the Spark DataFrame level.
HBaseCon2017 Efficient and portable data processing with Apache Beam and HBaseHBaseCon
In this talk we introduce Apache Beam, a unified model to create efficient and portable data processing pipelines. Beam uses a single set of abstractions to implement both batch and streaming computations that can be executed in different environments, e.g. Apache Spark, Apache Flink and Google Dataflow. Beam not only does data processing, but can be used as a tool to ingest/extract data to/from different data stores including HBase. We will present interaction scenarios between HBase and Beam and explore Beam's Input/Output (IO) model and how we leverage it to provide support for HBase.
Our team is responsible for storage at Xiaomi and we provide storage services for dozens of businesses, such as personal cloud storage for smart phones and user profile data. So we will share some practices and improvements of HBase at Xiaomi:
1: We upgraded most of our cluster from 0.94 to 0.98 in the last year and will share some experience about upgrading.
2: We encountered some problems and made some improvements on replication.
3: We fixed or still fixing some confusing behavior from client side.
4: We introduced some improvements on scan to make users easy to use and reduce the time of RPC requests.
5: We implement an asynchronous hbase client which is an important feature for HBase 2.0.
This document summarizes Salesforce's use of HBase and Phoenix for storing and querying large amounts of unstructured data at scale. Some key details include:
- Salesforce uses over 100 HBase clusters to store both customer and internal data, handling over 4 billion write requests and 600 million read requests per day.
- This includes storing login data, archived relational data, user activity, machine metrics and more, totaling over 80 terabytes written and 500 gigabytes read daily.
- An internal metrics database collects data from over 80,000 machines, storing 11.4 trillion metrics and growing, with 2.8 trillion metrics added in the last 6 months alone.
HBaseCon2017 Community-Driven Graphs with JanusGraphHBaseCon
Graphs are well-suited for many use cases to express and process complex relationships among entities in enterprise and social contexts. Fueled by the growing interest in graphs, there are various graph databases and processing systems that dot the graph landscape. JanusGraph is a community-driven project that continues the legacy of Titan, a pioneer of open source graph databases. JanusGraph is a scalable graph database optimized for large scale transactional and analytical graph processing. In the session, we will introduce JanusGraph, which features full integration with the Apache TinkerPop graph stack. We will discuss JanusGraph's optimized storage model that relies on HBase for fast graph transversal and processing.
by Jason Plurad and Jing Chen He of IBM
Redefining Cybersecurity with AI CapabilitiesPriyanka Aash
In this comprehensive overview of Cisco's latest innovations in cybersecurity, the focus is squarely on resilience and adaptation in the face of evolving threats. The discussion covers the imperative of tackling Mal information, the increasing sophistication of insider attacks, and the expanding attack surfaces in a hybrid work environment. Emphasizing a shift towards integrated platforms over fragmented tools, Cisco introduces its Security Cloud, designed to provide end-to-end visibility and robust protection across user interactions, cloud environments, and breaches. AI emerges as a pivotal tool, from enhancing user experiences to predicting and defending against cyber threats. The blog underscores Cisco's commitment to simplifying security stacks while ensuring efficacy and economic feasibility, making a compelling case for their platform approach in safeguarding digital landscapes.
DefCamp_2016_Chemerkin_Yury-publish.pdf - Presentation by Yury Chemerkin at DefCamp 2016 discussing mobile app vulnerabilities, data protection issues, and analysis of security levels across different types of mobile applications.
Retrieval Augmented Generation Evaluation with RagasZilliz
Retrieval Augmented Generation (RAG) enhances chatbots by incorporating custom data in the prompt. Using large language models (LLMs) as judge has gained prominence in modern RAG systems. This talk will demo Ragas, an open-source automation tool for RAG evaluations. Christy will talk about and demo evaluating a RAG pipeline using Milvus and RAG metrics like context F1-score and answer correctness.
The Challenge of Interpretability in Generative AI Models.pdfSara Kroft
Navigating the intricacies of generative AI models reveals a pressing challenge: interpretability. Our blog delves into the complexities of understanding how these advanced models make decisions, shedding light on the mechanisms behind their outputs. Explore the latest research, practical implications, and ethical considerations, as we unravel the opaque processes that drive generative AI. Join us in this insightful journey to demystify the black box of artificial intelligence.
Dive into the complexities of generative AI with our blog on interpretability. Find out why making AI models understandable is key to trust and ethical use and discover current efforts to tackle this big challenge.
Finetuning GenAI For Hacking and DefendingPriyanka Aash
Generative AI, particularly through the lens of large language models (LLMs), represents a transformative leap in artificial intelligence. With advancements that have fundamentally altered our approach to AI, understanding and leveraging these technologies is crucial for innovators and practitioners alike. This comprehensive exploration delves into the intricacies of GenAI, from its foundational principles and historical evolution to its practical applications in security and beyond.
"Making .NET Application Even Faster", Sergey Teplyakov.pptxFwdays
In this talk we're going to explore performance improvement lifecycle, starting with setting the performance goals, using profilers to figure out the bottle necks, making a fix and validating that the fix works by benchmarking it. The talk will be useful for novice and seasoned .NET developers and architects interested in making their application fast and understanding how things work under the hood.
Increase Quality with User Access Policies - July 2024Peter Caitens
⭐️ Increase Quality with User Access Policies ⭐️, presented by Peter Caitens and Adam Best of Salesforce. View the slides from this session to hear all about “User Access Policies” and how they can help you onboard users faster with greater quality.
Demystifying Neural Networks And Building Cybersecurity ApplicationsPriyanka Aash
In today's rapidly evolving technological landscape, Artificial Neural Networks (ANNs) have emerged as a cornerstone of artificial intelligence, revolutionizing various fields including cybersecurity. Inspired by the intricacies of the human brain, ANNs have a rich history and a complex structure that enables them to learn and make decisions. This blog aims to unravel the mysteries of neural networks, explore their mathematical foundations, and demonstrate their practical applications, particularly in building robust malware detection systems using Convolutional Neural Networks (CNNs).
This PDF delves into the aspects of information security from a forensic perspective, focusing on privacy leaks. It provides insights into the methods and tools used in forensic investigations to uncover and mitigate privacy breaches in mobile and cloud environments.
Exchange, Entra ID, Conectores, RAML: Todo, a la vez, en todas partes
hbaseconasia2017: Large scale data near-line loading method and architecture
1. Large scale data near-line loading
method and architecture
FiberHome Telecommunication
2017-7-19
2. /usr/bin/whoami
Shuaifeng Zhou(周帅锋):
• Big data research and development director ( Fiberhome 2013-)
• Software engineer (Huawei 2007-2013)
• Use and contribute to HBase since 2009
• sfzhou1791@fiberhome.com
4. HBase Realtime Data Loading
WAL/Flush/Compact
Triple IO pressure
Read/Write operations
share resource:
Cpu
Network
Disk IO
Handler
Read performance
decrease too much when
write load is heavy
5. Why near-line data loading?
DelayScale
ReliableResource
Large scale data loading reliably with acceptable time delay and resource occupation
Billions write ops per
region server one day
Usually, several minutes
delay is acceptable for
customers
Resource occupied can
be limited under an
acceptable level
Write op can be
repeated
Optimistic failure
handling
HBase
7. Read-Write split data loading
Independent WriterServer to
handle put request
RegionServer only handle
read request
WriteServer write HFile on
HDFS, send do-bulkload
operation.
Several minutes delay
between put and data
readable.
9. WriteServer Master
Task Management
• Create new loading tasks
every five minutes or every
10,000 records
• Find a slave to load the task
• Task status control
Topic Management
• Discover new kafka topics
• Receive loading request
• Loading records statistic
Slaves Management
• Slave status report to
master
• Balance
• failover
11. Failure Handling
Meta Data
based
Failure
Handling
Recover: Redo failed tasks when slave
down or master restart.
Task Meta Data is constructed when a
task is created by master, and change
status to succeed when slave finish the
task.
Task Meta Data is the descricption info of
a task, include the topic, partitions, start
and end offset, status. Stored on disk.
13. Balance
Load balance according tasks:
• Send new tasks to slaves with less tasks on handling
• Try to send tasks of one topic to a few fixed slaves
−avoid one region open everywhere
−Less region open, less small files
• Keep region opened for a while, even there are no tasks
− avoid region open/close too frequently
14. Compact
2
1
3 4 6
5
7 8
9
• Small files with higher priority
• Avoid one large file together with many small
files compact again and again
compact compact
15. StoreEngine
Customized store engine:
• organize store files in two queues
−one can be read and compact
−The other can only be compact
−If there are too many files, new file will not be readable
until they are compact
• Some new files discovered later better than all files can not
be read before time out
− Occasionally data explosion can be handled
− Region need split
− “Hot key” should be handled
16. HDFS Heterogeneous Storage Usage
• Use SSD storage as WriteServer tmp dir
• Use SATA as HBase data dir storage
− WriteServer write HFile on SSD
− Load HFile to HBase(Only move)
− Change to SATA storage after compact by regionServer
HDFS
SSD Storage SATA Storagecompact
WriteServer RegionServer
17. Resource Control
Resource used by WriteServer should be
controllable:
• Memory:
−JVM parameters 30~50GB memory
−Large Memory Store will avoid small files
−Too Large memory store will cause gc problems
• CPU:
− Slave can use 80% cpu cores at most
− Compare to real-time data load, a big optimize is we can control
the cpu occupation by write operations.
22. Summarize
We proposed an read-write split near-line
loading method and architecture:
• Increase loading performance
• Control resource used by write operation, make sure read
operation can not be starved
• Provide an architecture corresponding with kafka and hdfs
• Provide some optimize method, eg: compact, balance, etc.
• Provide test result