Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a herd of language models that natively support multilinguality, coding, reasoning, and tool usage. Our largest model is a dense Transformer with 405B parameters and a context window of up to 128K tokens. This paper presents an extensive empirical evaluation of Llama 3. We find that Llama 3 delivers comparable quality to leading language models such as GPT-4 on a plethora of tasks. We publicly release Llama 3, including pre-trained and post-trained versions of the 405B parameter language model and our Llama Guard 3 model for input and output safety. The paper also presents the results of experiments in which we integrate image, video, and speech capabilities into Llama 3 via a compositional approach. We observe this approach performs competitively with the state-of-the-art on image, video, and speech recognition tasks. The resulting models are not yet being broadly released as they are still under development.

How Facebook is analyzing big data.

The size of data sets being collected and analyzed in the industry for business intelligence is growing rapidly, making traditional warehousing solutions prohibitively expensive. Hadoop [1] is a popular open-source map-reduce implementation which is being used in companies like Yahoo, Facebook etc. to store and process extremely large data sets on commodity hardware. However, the map-reduce programming model is very low level and requires developers to write custom programs which are hard to maintain and reuse. In this paper, we present Hive, an open-source data warehousing solution built on top of Hadoop. Hive supports queries expressed in a SQL-like declarative language - HiveQL, which are compiled into map-reduce jobs that are executed using Hadoop. In addition, HiveQL enables users to plug in custom map-reduce scripts into queries. The language includes a type system with support for tables containing primitive types, collections like arrays and maps, and nested compositions of the same. The underlying IO libraries can be extended to query data in custom formats. Hive also includes a system catalog - Metastore - that contains schemas and statistics, which are useful in data exploration, query optimization and query compilation. In Facebook, the Hive warehouse contains tens of thousands of tables and stores over 700TB of data and is being used extensively for both reporting and ad-hoc analyses by more than 200 users per month.

Scalable analysis on large data sets has been core to the functions of a number of teams at Facebook - both engineering and non-engineering. Apart from ad hoc analysis of data and creation of business intelligence dashboards by analysts across the company, a number of Facebook's site features are also based on analyzing large data sets. These features range from simple reporting applications like Insights for the Facebook Advertisers, to more advanced kinds such as friend recommendations. In order to support this diversity of use cases on the ever increasing amount of data, a flexible infrastructure that scales up in a cost effective manner, is critical. We have leveraged, authored and contributed to a number of open source technologies in order to address these requirements at Facebook. These include Scribe, Hadoop and Hive which together form the cornerstones of the log collection, storage and analytics infrastructure at Facebook. In this paper we will present how these systems have come together and enabled us to implement a data warehouse that stores more than 15PB of data (2.5PB after compression) and loads more than 60TB of new data (10TB after compression) every day. We discuss the motivations behind our design choices, the capabilities of this solution, the challenges that we face in day today operations and future capabilities and improvements that we are working on.

We describe how aggregation is handled in the Trio system for uncertain and probabilistic data. Because “exact” aggregation in uncertain databases can produce exponentially sized results, we provide three alternatives: a low bound on the aggregate value, a high bound on the value, and the expected value. These variants return a single result instead of a set of possible results, and they are generally efficient to compute for both full-table and grouped aggregation queries. We provide formal definitions and semantics and a description of our open source implementation for single-table aggregation queries. We study the performance and scalability of our algorithms through experiments over a large synthetic data set. We also provide some preliminary results on aggregations over joins.

The size of data sets being collected and analyzed in the industry for business intelligence is growing rapidly, making traditional warehousing solutions prohibitively expensive. Hadoop [3] is a popular open-source map-reduce implementation which is being used as an alternative to store and process extremely large data sets on commodity hardware. However, the map-reduce programming model is very low level and requires developers to write custom programs which are hard to maintain and reuse.

Trio is a new kind of database system that supports data, uncertainty, and lineage in a fully integrated manner. The first Trio prototype, dubbed Trio-One, is built on top of a conventional DBMS using data and query translation techniques together with a small number of stored procedures. This paper describes Trio-One's translation scheme and system architecture, showing how it efficiently and easily supports the Trio data model and query language.