DOCSLIB.ORG

Combining ACID and BASE! in a Distributed Database

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Combining ACID and BASE! in a Distributed Database Salt Combining ACID and BASE! in a distributed database Chao Xie, Chunzhi Su, Manos Kapritsos, Yang Wang,! Navid Yaghmazadeh, Lorenzo Alvisi, Prince Mahajan The University of Texas at Austin TRANSACTIONS ARE GREAT Four properties in a single abstraction Atomicity Consistency Isolation Durability (ACID) • Ease of programming • Easy to reason about Transaction TRANSACTIONS ARE GREAT Txn Manager Data Nodes Concurrency control! Write(x) Phase 1 limits performance Write(y) Commit 2PC protocol is costly Phase 2 THE ALTERNATIVE: BASE Application Storage Interface Transaction ACID Storage Guarantees THE ALTERNATIVE: BASE • Write custom code to get better performance Application Storage Interface BASE Storage (e.g., put, get) THE ALTERNATIVE: BASE • Write custom code to get better performance • Complexity gets out of control Implement ApplicationApplicationConsistency Storage Interface BASE Storage (e.g., put, get) A STARK CHOICE Ease of programming & Performance Performance BASE ACID Ease of programming 20% of the causes! account for! 80% of the effects Vilfredo Pareto NOT ALL TRANSACTIONS ARE CREATED EQUAL 20% of the causes! account for! 80% of the effects •Many transactions are not run frequently •Many transactions are lightweight AN OPPORTUNITY Txn 1 Txn 2 BASE •Identify critical transactions •BASE-ify only critical transactions SALT Motivation! Base Transactions & Salt Isolation! Achieving Salt Isolation! Evaluation MORE CONCURRENCY ! Transfer Part1 Transfer Is c≥$10? Is c≥$10? c=c-$10 c=c-$10 Part2 s=s+$10 s=s+$10 MORE CONCURRENCY ! Time Transfer Transfer Is c≥$10? Part1 c=c-$10 Is c≥$10? s=s+$10 c=c-$10 Transfer Transfer Part2 Part1 Is c≥$10? Is c≥$10? s=s+$10 c=c-$10 c=c-$10 s=s+$10 Part2 s=s+$10 CORRECTNESS AT RISK Transfer Balance Is c≥$10? Read c c=c-$10 Read s s=s+$10 CORRECTNESS AT RISK Transfer Part1 Is c≥$10? Balance c=c-$10 Exposed ! Read c state Read s Part2 s=s+$10 CORRECTNESS AT RISK Part1 Is c≥$10? c=c-$10 Balance Finer Isolation for one transaction! Read c may affect all transactions!! Read s Part2 s=s+$10 Performance vs Complexity Better Performance More Interleavings! More Complexity Performance vs Complexity Better Performance More Interleavings ! (only among perf-critical txns) Other Transactions Unaffected Transfer 1 Transfer 2 Is c≥$10? Is c≥$10? Balance c=c-$10 c=c-$10 Read c Read s s=s+$10 s=s+$10 Behaves differently when interacting with different transactions Time Transfer 1 Is c≥$10? c=c-$10 Transfer 2 Is c≥$10? s=s+$10 c=c-$10 s=s+$10 Balance Read c Read s BASE TRANSACTION BASE transaction alkaline txn alkaline txn Behaves differently when interacting with different transactions BASE INTERACT WITH BASE BASE BASE 1 2 Fine Isolation granularity! between BASE transactions BASE INTERACT WITH BASE Fine Isolation granularity! between BASE transactions BASE INTERACT WITH ACID BASE ACID BASE transactions provide coarse Isolation! granularity to ACID transactions BASE INTERACT WITH ACID ACID BASE BASE transactions provide coarse Isolation! granularity to ACID transactions SALT ISOLATION BASE transactions: multiple granularities of Isolation To BASE transactions:! To ACID transactions:! a sequence of small ! a single, monolithic ! ACID transactions ACID transaction Performance & Ease of Programming SALT Motivation! Base Transactions & Salt Isolation! Achieving Salt Isolation! Evaluation ONE MECHANISM LOCKS Three flavors ACID locks! Alkaline locks! Saline locks ACID LOCKS ACID 1 AC-R AC-W Write x AC-R √ X ACID 2 AC-W X X Read x Lock Table Execute LOCKS? BASE Wx BASE Rx Execute ACID Rx Execute LOCKS? BASE alkaline lock Wx BASE saline lock Rx Execute ACID Rx Execute ALKALINE LOCKS alkaline lock BASE BASE ACID AC-R AC-W alk-R alk-W Wx Rx Rx AC-R √ X √ X AC-W X X X X Wait Wait alk-R √ X √ X alk-W X X X X Lock Table Conflict with ACID & alkaline locks SALINE LOCKS saline lock BASE AC-R AC-W alk-R alk-W sal-R sal-W AC-R √ X √ X √ X Wx AC-W X X X X X X alk-R √ X √ X √ √ ACID alk-W X X X X √ √ Rx sal-R √ X √ √ √ √ sal-W X X √ √ √ √ Wait Lock Table Conflict only with ACID locks A SUBTLE PROBLEM BASE W x BASE R x y=x ACID R y Dirty Read! ACID reads uncommitted value of x! A SUBTLE PROBLEM BASE W x BASE R x y=x ACID R y Dirty Read! For the solution, please read our paper THE BOTTOM LINE Guarantee! Salt prevents all ACID transactions from being affected by BASE transactions either directly or indirectly. SALT Motivation! Base Transactions & Salt Isolation! Achieving Salt Isolation! Evaluation QUESTIONS TO ANSWER What is the performance gain of Salt compared to ACID? Can we get most performance gain compared to the BASE approach? EXPERIMENTAL SETUP Configuration! • Emulab Cluster (Dell Power Edge R710)! • 10 shards, 3-way replicated! Workloads! • TPC-C! • Fusion Ticket! • Microbenchmarks PERFORMANCE GAIN Fusion Ticket 250 200 Salt 150 ACID 100 Latency (ms) Latency 6.5X 50 0 0 1000 2000 3000 4000 5000 6000 7000 8000 Throughput (transactions/sec) REAP MOST PERFORMANCE OF BASE Fusion Ticket 10000 8000 6000 6.5X 7.2X 4000 … … 2000 0 Throughput(transactions/sec) ACID 1 2 3 Raw ops Number of BASE-ified transactions RELATED WORK Optimizing ACID Performance! • H-Store, Granola, F1, Sagas, Transaction Chain, Calvin …! ! BASE with enhanced semantics (e.g., partition local transactions)! • ElasTras, G-Store, Megastore … SALT Pain Point Key Abstraction! Promising! Transactional systems ! Base Transaction do not scale Results.
Load more

Recommended publications
  • SD-SQL Server: Scalable Distributed Database System
    The International Arab Journal of Information Technology, Vol. 4, No. 2, April 2007 103 SD -SQL Server: Scalable Distributed Database System Soror Sahri CERIA, Université Paris Dauphine, France Abstract: We present SD -SQL Server, a prototype scalable distributed database system. It let a relational table to grow over new storage nodes invis ibly to the application. The evolution uses splits dynamically generating a distributed range partitioning of the table. The splits avoid the reorganization of a growing database, necessary for the current DBMSs and a headache for the administrators. We il lustrate the architecture of our system, its capabilities and performance. The experiments with the well -known SkyServer database show that the overhead of the scalable distributed table management is typically minimal. To our best knowledge, SD -SQL Server is the only DBMS with the discussed capabilities at present. Keywords: Scalable distributed DBS, scalable table, distributed partitioned view, SDDS, performance . Received June 4 , 2005; accepted June 27 , 2006 1. Introduction or k -d based with respect to the partitioning key(s). The application sees a scal able table through a specific The explosive growth of the v olume of data to store in type of updateable distributed view termed (client) databases makes many of them huge and permanently scalable view. Such a view hides the partitioning and growing. Large tables have to be hash ed or partitioned dynamically adjusts itself to the partitioning evolution. over several storage sites. Current DBMSs, e. g., SQL The adjustment is lazy, in the sense it occurs only Server, Oracle or DB2 to name only a few, provide when a query to the scalable table comes in and the static partitioning only.
    [Show full text]
  • Socrates: the New SQL Server in the Cloud
    Socrates: The New SQL Server in the Cloud Panagiotis Antonopoulos, Alex Budovski, Cristian Diaconu, Alejandro Hernandez Saenz, Jack Hu, Hanuma Kodavalla, Donald Kossmann, Sandeep Lingam, Umar Farooq Minhas, Naveen Prakash, Vijendra Purohit, Hugh Qu, Chaitanya Sreenivas Ravella, Krystyna Reisteter, Sheetal Shrotri, Dixin Tang, Vikram Wakade Microsoft Azure & Microsoft Research ABSTRACT 1 INTRODUCTION The database-as-a-service paradigm in the cloud (DBaaS) The cloud is here to stay. Most start-ups are cloud-native. is becoming increasingly popular. Organizations adopt this Furthermore, many large enterprises are moving their data paradigm because they expect higher security, higher avail- and workloads into the cloud. The main reasons to move ability, and lower and more flexible cost with high perfor- into the cloud are security, time-to-market, and a more flexi- mance. It has become clear, however, that these expectations ble “pay-as-you-go” cost model which avoids overpaying for cannot be met in the cloud with the traditional, monolithic under-utilized machines. While all these reasons are com- database architecture. This paper presents a novel DBaaS pelling, the expectation is that a database runs in the cloud architecture, called Socrates. Socrates has been implemented at least as well as (if not better) than on premise. Specifically, in Microsoft SQL Server and is available in Azure as SQL DB customers expect a “database-as-a-service” to be highly avail- Hyperscale. This paper describes the key ideas and features able (e.g., 99.999% availability), support large databases (e.g., of Socrates, and it compares the performance of Socrates a 100TB OLTP database), and be highly performant.
    [Show full text]
  • An Overview of Distributed Databases
    International Journal of Information and Computation Technology. ISSN 0974-2239 Volume 4, Number 2 (2014), pp. 207-214 © International Research Publications House http://www. irphouse.com /ijict.htm An Overview of Distributed Databases Parul Tomar1 and Megha2 1Department of Computer Engineering, YMCA University of Science & Technology, Faridabad, INDIA. 2Student Department of Computer Science, YMCA University of Science and Technology, Faridabad, INDIA. Abstract A Database is a collection of data describing the activities of one or more related organizations with a specific well defined structure and purpose. A Database is controlled by Database Management System(DBMS) by maintaining and utilizing large collections of data. A Distributed System is the one in which hardware and software components at networked computers communicate and coordinate their activity only by passing messages. In short a Distributed database is a collection of databases that can be stored at different computer network sites. This paper presents an overview of Distributed Database System along with their advantages and disadvantages. This paper also provides various aspects like replication, fragmentation and various problems that can be faced in distributed database systems. Keywords: Database, Deadlock, Distributed Database Management System, Fragmentation, Replication. 1. Introduction A Database is systematically organized or structuredrepository of indexed information that allows easy retrieval, updating, analysis, and output of data. Each database may involve different database management systems and different architectures that distribute the execution of transactions [1]. A distributed database is a database in which storage devices are not all attached to a common processing unit such as the CPU. It may be stored in multiple computers, located in the same physical location; or may be dispersed over a network of interconnected computers.
    [Show full text]
  • Blockchain Database for a Cyber Security Learning System
    Session ETD 475 Blockchain Database for a Cyber Security Learning System Sophia Armstrong Department of Computer Science, College of Engineering and Technology East Carolina University Te-Shun Chou Department of Technology Systems, College of Engineering and Technology East Carolina University John Jones College of Engineering and Technology East Carolina University Abstract Our cyber security learning system involves an interactive environment for students to practice executing different attack and defense techniques relating to cyber security concepts. We intend to use a blockchain database to secure data from this learning system. The data being secured are students’ scores accumulated by successful attacks or defends from the other students’ implementations. As more professionals are departing from traditional relational databases, the enthusiasm around distributed ledger databases is growing, specifically blockchain. With many available platforms applying blockchain structures, it is important to understand how this emerging technology is being used, with the goal of utilizing this technology for our learning system. In order to successfully secure the data and ensure it is tamper resistant, an investigation of blockchain technology use cases must be conducted. In addition, this paper defined the primary characteristics of the emerging distributed ledgers or blockchain technology, to ensure we effectively harness this technology to secure our data. Moreover, we explored using a blockchain database for our data. 1. Introduction New buzz words are constantly surfacing in the ever evolving field of computer science, so it is critical to distinguish the difference between temporary fads and new evolutionary technology. Blockchain is one of the newest and most developmental technologies currently drawing interest.
    [Show full text]
  • Guide to Design, Implementation and Management of Distributed Databases
    NATL INST OF STAND 4 TECH H.I C NIST Special Publication 500-185 A111D3 MTfiDfi2 Computer Systems Guide to Design, Technology Implementation and Management U.S. DEPARTMENT OF COMMERCE National Institute of of Distributed Databases Standards and Technology Elizabeth N. Fong Nisr Charles L. Sheppard Kathryn A. Harvill NIST I PUBLICATIONS I 100 .U57 500-185 1991 C.2 NIST Special Publication 500-185 ^/c 5oo-n Guide to Design, Implementation and Management of Distributed Databases Elizabeth N. Fong Charles L. Sheppard Kathryn A. Harvill Computer Systems Laboratory National Institute of Standards and Technology Gaithersburg, MD 20899 February 1991 U.S. DEPARTMENT OF COMMERCE Robert A. Mosbacher, Secretary NATIONAL INSTITUTE OF STANDARDS AND TECHNOLOGY John W. Lyons, Director Reports on Computer Systems Technology The National institute of Standards and Technology (NIST) has a unique responsibility for computer systems technology within the Federal government, NIST's Computer Systems Laboratory (CSL) devel- ops standards and guidelines, provides technical assistance, and conducts research for computers and related telecommunications systems to achieve more effective utilization of Federal Information technol- ogy resources. CSL's responsibilities include development of technical, management, physical, and ad- ministrative standards and guidelines for the cost-effective security and privacy of sensitive unclassified information processed in Federal computers. CSL assists agencies in developing security plans and in improving computer security awareness training. This Special Publication 500 series reports CSL re- search and guidelines to Federal agencies as well as to organizations in industry, government, and academia. National Institute of Standards and Technology Special Publication 500-185 Natl. Inst. Stand. Technol.
    [Show full text]
  • A Transaction Processing Method for Distributed Database
    Advances in Computer Science Research, volume 87 3rd International Conference on Mechatronics Engineering and Information Technology (ICMEIT 2019) A Transaction Processing Method for Distributed Database Zhian Lin a, Chi Zhang b School of Computer and Cyberspace Security, Communication University of China, Beijing, China [email protected], [email protected] Abstract. This paper introduces the distributed transaction processing model and two-phase commit protocol, and analyses the shortcomings of the two-phase commit protocol. And then we proposed a new distributed transaction processing method which adds heartbeat mechanism into the two- phase commit protocol. Using the method can improve reliability and reduce blocking in distributed transaction processing. Keywords: distributed transaction, two-phase commit protocol, heartbeat mechanism. 1. Introduction Most database services of application systems will be distributed on several servers, especially in some large-scale systems. Distributed transaction processing will be involved in the execution of business logic. At present, two-phase commit protocol is one of the methods to distributed transaction processing in distributed database systems. The two-phase commit protocol includes coordinator (transaction manager) and several participants (databases). In the process of communication between the coordinator and the participants, if the participants without reply for fail, the coordinator can only wait all the time, which can easily cause system blocking. In this paper, heartbeat mechanism is introduced to monitor participants, which avoid the risk of blocking of two-phase commit protocol, and improve the reliability and efficiency of distributed database system. 2. Distributed Transactions 2.1 Distributed Transaction Processing Model In a distributed system, each node is physically independent and they communicates and coordinates each other through the network.
    [Show full text]
  • Implementing Distributed Transactions Distributed Transaction Distributed Database Systems ACID Properties Global Atomicity Atom
    Distributed Transaction • A distributed transaction accesses resource managers distributed across a network Implementing Distributed • When resource managers are DBMSs we refer to the Transactions system as a distributed database system Chapter 24 DBMS at Site 1 Application Program DBMS 1 at Site 2 2 Distributed Database Systems ACID Properties • Each local DBMS might export • Each local DBMS – stored procedures, or – supports ACID properties locally for each subtransaction – an SQL interface. • Just like any other transaction that executes there • In either case, operations at each site are grouped – eliminates local deadlocks together as a subtransaction and the site is referred • The additional issues are: to as a cohort of the distributed transaction – Global atomicity: all cohorts must abort or all commit – Each subtransaction is treated as a transaction at its site – Global deadlocks: there must be no deadlocks involving • Coordinator module (part of TP monitor) supports multiple sites ACID properties of distributed transaction – Global serialization: distributed transaction must be globally serializable – Transaction manager acts as coordinator 3 4 Atomic Commit Protocol Global Atomicity Transaction (3) xa_reg • All subtransactions of a distributed transaction Manager Resource must commit or all must abort (coordinator) Manager (1) tx_begin (cohort) • An atomic commit protocol, initiated by a (4) tx_commit (5) atomic coordinator (e.g., the transaction manager), commit protocol ensures this. (3) xa_reg Resource Application – Coordinator
    [Show full text]
  • Intelligent Implementation Processor Design for Oracle Distributed Databases System
    International Conference on Control, Engineering & Information Technology (CEIT’14) Proceedings - Copyright IPCO-2014, pp. 278-296 ISSN 2356-5608 Intelligent Implementation Processor Design for Oracle Distributed Databases System Hassen Fadoua, Grissa Touzi Amel 1Université Tunis El Manar, LIPAH, FST, Tunisia 2Université Tunis El Manar, ENIT, LIPAH,FST, Tunisia {[email protected];[email protected]} Abstract. Despite the increasing need for modeling and implementing Distributed Databases (DDB), distributed database management systems are still quite far from helping the designer to directly implement its BDD. Indeed, the fundamental principle of implementation of a DDB is to make the database appear as a centralized database, providing series of transparencies, something that is not provided directly by the current DDBMS. We focus in this work on Oracle DBMS which, despite its market dominance, offers only a few logical mechanisms to implement distribution. To remedy this problem, we propose a new architecture of DDBMS Oracle. The idea is based on extending it by an intelligent layer that provides: 1) creation of different types of fragmentation through a GUI for defining different sites geographically dispersed 2) allocation and replication of DB. The system must automatically generate SQL scripts for each site of the original configuration. Keywords : distributed databases; big data; fragmentation; allocation; replication. 1 Introduction The organizational evolution of companies and institutions that rely on computer systems to manage their data has always been hampered by centralized structures already installed, this architecture does not respond to the need for autonomy and evolution of the organization because it requires a permanent return to the central authority, which leads to a huge waste of time and an overwhelmed work.
    [Show full text]
  • Distributed Databases What Is a Distributed Database?
    Distributed Databases Chapter 16 1 What is a Distributed Database? • Database whose relations reside on different sites • Database some of whose relations are replicated at different sites • Database whose relations are split between different sites 2 1 Two Types of Applications that Access Distributed Databases • The application accesses data at the level of SQL statements – Example: company has nationwide network of warehouses, each with its own database; a transaction can access all databases using their schemas • The application accesses data at a database using only stored procedures provided by that database. – Example: purchase transaction involving a merchant and a credit card company, each providing stored subroutines for its subtransactions 3 Optimizing Distributed Queries • Only applications of the first type can access data directly and hence employ query optimization strategies • These are the applications we consider in this chapter 4 2 Some Issues · How should a distributed database be designed? · At what site should each item be stored? · Which items should be replicated and at which sites? · How should queries that access multiple databases be processed? · How do issues of query optimization affect query design? 5 Why Might Data Be Distributed · Data might be distributed to minimize communication costs or response time · Data might be kept at the site where it was created so that its creators can maintain control and security · Data might be replicated to increase its availability in the event of failure or to decrease
    [Show full text]
  • Salt: Combining ACID and BASE in a Distributed Database
    Salt: Combining ACID and BASE in a Distributed Database Chao Xie, Chunzhi Su, Manos Kapritsos, Yang Wang, Navid Yaghmazadeh, Lorenzo Alvisi, and Prince Mahajan, The University of Texas at Austin https://www.usenix.org/conference/osdi14/technical-sessions/presentation/xie This paper is included in the Proceedings of the 11th USENIX Symposium on Operating Systems Design and Implementation. October 6–8, 2014 • Broomfield, CO 978-1-931971-16-4 Open access to the Proceedings of the 11th USENIX Symposium on Operating Systems Design and Implementation is sponsored by USENIX. Salt: Combining ACID and BASE in a Distributed Database Chao Xie, Chunzhi Su, Manos Kapritsos, Yang Wang, Navid Yaghmazadeh, Lorenzo Alvisi, Prince Mahajan The University of Texas at Austin Abstract: This paper presents Salt, a distributed recently popularized by several NoSQL systems [1, 15, database that allows developers to improve the perfor- 20, 21, 27, 34]. Unlike ACID, BASE offers more of a set mance and scalability of their ACID applications through of programming guidelines (such as the use of parti- the incremental adoption of the BASE approach. Salt’s tion local transactions [32, 37]) than a set of rigorously motivation is rooted in the Pareto principle: for many ap- specified properties and its instantiations take a vari- plications, the transactions that actually test the perfor- ety of application-specific forms. Common among them, mance limits of ACID are few. To leverage this insight, however, is a programming style that avoids distributed Salt introduces BASE transactions, a new abstraction transactions to eliminate the performance and availabil- that encapsulates the workflow of performance-critical ity costs of the associated distributed commit protocol.
    [Show full text]
  • An Evaluation of Distributed Concurrency Control
    An Evaluation of Distributed Concurrency Control Rachael Harding Dana Van Aken MIT CSAIL Carnegie Mellon University [email protected] [email protected] Andrew Pavlo Michael Stonebraker Carnegie Mellon University MIT CSAIL [email protected] [email protected] ABSTRACT there is little understanding of the trade-offs in a modern cloud Increasing transaction volumes have led to a resurgence of interest computing environment offering high scalability and elasticity. Few in distributed transaction processing. In particular, partitioning data of the recent publications that propose new distributed protocols across several servers can improve throughput by allowing servers compare more than one other approach. For example, none of the to process transactions in parallel. But executing transactions across papers published since 2012 in Table 1 compare against timestamp- servers limits the scalability and performance of these systems. based or multi-version protocols, and seven of them do not compare In this paper, we quantify the effects of distribution on concur- to any other serializable protocol. As a result, it is difficult to rency control protocols in a distributed environment. We evaluate six compare proposed protocols, especially as hardware and workload classic and modern protocols in an in-memory distributed database configurations vary across publications. evaluation framework called Deneva, providing an apples-to-apples Our aim is to quantify and compare existing distributed concur- comparison between each. Our results expose severe limitations of rency control protocols for in-memory DBMSs. We develop an distributed transaction processing engines. Moreover, in our anal- empirical understanding of the behavior of distributed transactions ysis, we identify several protocol-specific scalability bottlenecks.
    [Show full text]
  • Trends in Development of Databases and Blockchain
    Trends in Development of Databases and Blockchain Mayank Raikwar∗, Danilo Gligoroski∗, Goran Velinov† ∗ Norwegian University of Science and Technology (NTNU) Trondheim, Norway † University Ss. Cyril and Methodius Skopje, Macedonia Email: {mayank.raikwar,danilog}@ntnu.no, goran.velinov@finki.ukim.mk Abstract—This work is about the mutual influence between some features which traditional database has. Blockchain can two technologies: Databases and Blockchain. It addresses two leverage the traditional database features by either integrat- questions: 1. How the database technology has influenced the ing the traditional database with blockchain or, to create a development of blockchain technology?, and 2. How blockchain technology has influenced the introduction of new functionalities blockchain-oriented distributed database. The inclusion of the in some modern databases? For the first question, we explain how database features will leverage the blockchain with low la- database technology contributes to blockchain technology by un- tency, high throughput, fast scalability, and complex queries on locking different features such as ACID (Atomicity, Consistency, blockchain data. Thus having the features of both blockchain Isolation, and Durability) transactional consistency, rich queries, and database, the application enhances its efficiency and real-time analytics, and low latency. We explain how the CAP (Consistency, Availability, Partition tolerance) theorem known security. Many of the blockchain platforms are now integrating for databases influenced
    [Show full text]