We consider information flow via physical transport of inscribed media through space and compare it to information flow via electromagnetic radiation. Somewhat counterintuitively for point to point links, physical transport of inscribed mass is often energetically more efficient by many orders of magnitude than electromagnetic broadcast. And perhaps more surprising, even in a broadcast setting ...

In [14], Lamport has defined three classes of shared registers which support read and write operations, called —safe, regular and atomic—depending on their properties when several reads and/or writes are executed concurrently. We consider generalizations of Lamport’s notions, called k-safe, k-regular and k-atomic. First, we provide constructions for implementing 1-atomic registers (the stronges...

Country N Main languages Argentina 12,450 Spanish Australia 83,531 English Austria 28,355 German Bangladesh 989 Bengali Belgium 41,411 French Dutch, German Brazil 28,152 Portuguese Bulgaria 3,370 Bulgarian Canada 145,630 French English China 42,511 Chinese Czechoslovakia 19,848 Czech, Slovak Denmark 39,018 Danish Egypt 5,163 Arabic Finland 34,783 Finnish France 186,848 French Germany 239,204 Ge...

The design of the logging and recovery components of database management systems (DBMSs) has always been influenced by the difference in the performance characteristics of volatile (DRAM) and non-volatile storage devices (HDD/SSDs). The key assumption has been that non-volatile storage is much slower than DRAM and only supports block-oriented read/writes. But the arrival of new nonvolatile memo...

Large writes are beneficial both on individual disks and on disk arrays, e.g., RAID-5. The presented design enables large writes of internal B-tree nodes and leaves. It supports both in-place updates and large append-only (“log-structured”) write operations within the same storage volume, within the same B-tree, and even at the same time. The essence of the proposal is to make page migration in...

We study the cost of using message passing to implement linearizable read write objects for shared memory multiprocessors under various assumptions on the available timing informa tion We take as cost measures the worst case response times for performing read and write operations in distributed implementations of virtual shared memory consisting of such objects and the sum of these response tim...