We live in the golden age of distributed computing. Public cloud platforms now offer virtually unlimited compute and storage resources on demand. At the same time, the Software-as-a-Service (SaaS) model brings enterprise-class systems to users who previously could not afford such systems due to their cost and complexity. Alas, traditional data warehousing systems are struggling to fit into this new environment. For one thing, they have been designed for fixed resources and are thus unable to leverage the cloud's elasticity. For another thing, their dependence on complex ETL pipelines and physical tuning is at odds with the flexibility and freshness requirements of the cloud's new types of semi-structured data and rapidly evolving workloads. We decided a fundamental redesign was in order. Our mission was to build an enterprise-ready data warehousing solution for the cloud. The result is the Snowflake Elastic Data Warehouse, or "Snowflake" for short. Snowflake is a multi-tenant, transactional, secure, highly scalable and elastic system with full SQL support and built-in extensions for semi-structured and schema-less data. The system is offered as a pay-as-you-go service in the Amazon cloud. Users upload their data to the cloud and can immediately manage and query it using familiar tools and interfaces. Implementation began in late 2012 and Snowflake has been generally available since June 2015. Today, Snowflake is used in production by a growing number of small and large organizations alike. The system runs several million queries per day over multiple petabytes of data.

The growth in compute speed has outpaced the growth in network bandwidth over the last decades. This has led to an increasing performance gap between local and distributed processing. A parallel database cluster thus has to maximize the locality of query processing. A common technique to this end is to co-partition relations to avoid expensive data shuffling across the network. However, this is limited to one attribute per relation and is expensive to maintain in the face of updates. Other attributes often exhibit a fuzzy co-location due to correlations with the distribution key but current approaches do not leverage this. In this paper, we introduce locality-sensitive data shuffling, which can dramatically reduce the amount of network communication for distributed operators such as join and aggregation. We present four novel techniques: (i) optimal partition assignment exploits locality to reduce the network phase duration; (ii) communication scheduling avoids bandwidth underutilization due to cross traffic; (iii) adaptive radix partitioning retains locality during data repartitioning and handles value skew gracefully; and (iv) selective broadcast reduces network communication in the presence of extreme value skew or large numbers of duplicates. We present comprehensive experimental results, which show that our techniques can improve performance by up to factor of 5 for fuzzy co-location and a factor of 3 for inputs with value skew.

An elastic and highly available data store is a key component of many cloud applications. Existing data stores with strong consistency guarantees are designed and optimized for small updates, key-value access, and (if supported) small range queries over a predefined key column. This raises performance and availability problems for applications which inherently require large updates, non-key access, and large range queries. This paper presents a solution to these problems: Crescando/RB; a distributed, scan-based, main memory, relational data store (single table) with robust performance and high availability. The system addresses a real, large-scale industry use case: the Amadeus travel management system. This paper focuses on the distribution layer of Crescando/RB, the problem and theory behind it, the rationale underlying key design decisions, and the novel multicast protocol and replication framework it is composed of. Highlighting the key features of the distribution layer, we present experimental results showing that even under permanent node failures and large-scale data repartitioning, Crescando/RB remains fully available and capable of sustaining a heavy query and update load.

This demonstration presents Crescando, an implementation of a distributed relational table that guarantees predictable response time on unpredictable workloads. In Crescando, data is stored in main memory and accessed via full-table scans. By using scans instead of index lookups, Crescando overcomes the read-write contention in index structures and eliminates the scalability issues that exist in traditional index-based systems. Crescando is specifically designed to process a large number of queries in parallel, allowing high query rates. The goal of this demonstration is to show the ability of Crescando to a) quickly answer arbitrary user-generated queries, and b) execute a large number of queries and updates in parallel, while providing strict response time and data freshness guarantees.

In this paper, we propose a distributed model for massive scalability in net-VEs. Our model inherits concepts from distributed databases, where the application logic and transaction processing take place at the client machine. The key feature of our model is its novel transaction model, which exploits application semantics to reduce the number of messages needed to maintain consistency. Our proposed model imposes no major restrictions on the interaction between participants located in different parts of the world, and so can be easily adapted to a wide range of net-VEs.

This paper introduces Crescando: a scalable, distributed relational table implementation designed to perform large numbers of queries and updates with guaranteed access latency and data freshness. To this end, Crescando leverages a number of modern query processing techniques and hardware trends. Specifically, Crescando is based on parallel, collaborative scans in main memory and so-called "query-data" joins known from data-stream processing. While the proposed approach is not always optimal for a given workload, it provides latency and freshness guarantees for all workloads. Thus, Crescando is particularly attractive if the workload is unknown, changing, or involves many different queries. This paper describes the design, algorithms, and implementation of a Crescando storage node, and assesses its performance on modern multi-core hardware.