Synthetic DNA is durable and can encode digital data with high density, making it an attractive medium for data storage. However, recovering stored data on a large-scale currently requires all the DNA in a pool to be sequenced, even if only a subset of the information needs to be extracted. Here, we encode and store 35 distinct files (over 200 MB of data), in more than 13 million DNA oligonucleotides, and show that we can recover each file individually and with no errors, using a random access approach. We design and validate a large library of primers that enable individual recovery of all files stored within the DNA. We also develop an algorithm that greatly reduces the sequencing read coverage required for error-free decoding by maximizing information from all sequence reads. These advances demonstrate a viable, large-scale system for DNA data storage and retrieval.

No abstract

This paper describes a learning-based approach to the acceleration of approximate programs. We describe the Parrot transformation, a program transformation that selects and trains a neural network to mimic a region of imperative code. After the learning phase, the compiler replaces the original code with an invocation of a low-power accelerator called a neural processing unit (NPU). The NPU is tightly coupled to the processor pipeline to accelerate small code regions. Since neural networks produce inherently approximate results, we define a programming model that allows programmers to identify approximable code regions-code that can produce imprecise but acceptable results. Offloading approximable code regions to NPUs is faster and more energy efficient than executing the original code. For a set of diverse applications, NPU acceleration provides wholeapplication speedup of 2.3× and energy savings of 3.0× on average with quality loss of at most 9.6%.

No abstract

No abstract

Transactional Memory (TM), Thread-Level Speculation (TLS), and Checkpointed multiprocessors are three popular architectural techniques based on the execution of multiple, cooperating speculative threads. In these environments, correctly maintaining data dependences across threads requires mechanisms for disambiguating addresses across threads, invalidating stale cache state, and making committed state visible. These mechanisms are both conceptually involved and hard to implement. In this paper, we present Bulk, a novel approach to simplify these mechanisms. The idea is to hash-encode a thread's access information in a concise signature, and then support in hardware signature operations that efficiently process sets of addresses. Such operations implement the mechanisms described. Bulk operations are inexact but correct, and provide substantial conceptual and implementation simplicity. We evaluate Bulk in the context of TLS using SPECint2000 codes and TM using multithreaded Java workloads. Despite its simplicity, Bulk has competitive performance with more complex schemes. We also find that signature configuration is a key design parameter.

Demand for data storage is growing exponentially, but the capacity of existing storage media is not keeping up. Using DNA to archive data is an attractive possibility because it is extremely dense, with a raw limit of 1 exabyte/mm 3 (10 9 GB/mm 3 ), and long-lasting, with observed half-life of over 500 years.This paper presents an architecture for a DNA-based archival storage system. It is structured as a key-value store, and leverages common biochemical techniques to provide random access. We also propose a new encoding scheme that offers controllable redundancy, trading off reliability for density. We demonstrate feasibility, random access, and robustness of the proposed encoding with wet lab experiments involving 151 kB of synthesized DNA and a 42 kB randomaccess subset, and simulation experiments of larger sets calibrated to the wet lab experiments. Finally, we highlight trends in biotechnology that indicate the impending practicality of DNA storage for much larger datasets.

No abstract