METHOD FOR PERFORMING IN-MEMORY PROCESSING OPERATIONS ON CONTINUOUS ALLOCATION DATA AND RELATED

Methods, apparatuses, and systems for in-memory or near-memory processing are described. A bit (e.g., vector) string may be extracted and processed in logic of the memory device without a separate processing unit. Operations (e.g., arithmetic operations) may be performed on numbers stored in bit ser...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: AKKAR AMEYA D, EILERT SEAN S, GUNASEKARAN, SRINIVASAN, YUDANOV DMITRY, PARTHASARATHY SIVAGNANAM
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Methods, apparatuses, and systems for in-memory or near-memory processing are described. A bit (e.g., vector) string may be extracted and processed in logic of the memory device without a separate processing unit. Operations (e.g., arithmetic operations) may be performed on numbers stored in bit serials during a single clock cycle sequence. Accordingly, arithmetic may be performed in a single pass, since the number is a bit of two or more bit strings, extraction does not need to store the number in the middle. Vectors may be extracted (e.g., identified, transmitted, received) from one or more bit lines. Register of the memory array may be used to write (e.g., store or temporarily store) results or auxiliary bits (e.g., carry bits or carry flags) that facilitate arithmetic operations. Circuitry near, adjacent to, or below the memory array may employ XOR or AND (or other) logic to extract, organize, or compute data. 描述用于存储器内或附近处理的方法、设备和系统。位(例如,向量)串可被提取并在存储器装置的逻辑中处理,而不需要单独的处理单元。可在单个时钟循环序列期间对以位串行方式存储的数字执行运算(例如,算术