Skewed Timestamps

In the well known timestamp algorithm each writing process reads all the registers of the other processes, and then sets a higher timestamp (natural number) to itself. In thèskewed algorithm' only one writing process in each pair reads the other. 1 Preface Messages are often timestamped. In a fax, the timestamp includes the date and exact time of the day, and in a book only the publication year, but in all cases this information guides the reader in choosing and processing the data. The antiquarian may choose the oldest book, and the student the newest edition, but in the general timestamp protocol the reader (called scanner) returns all the messages in their issuing order. Neither clocks nor calendars are needed in the Unbounded Timestamp Algorithm (Figure 1) which uses the natural numbers as timestamps. Each writer (called in this protocol Labeler) rst reads the registers of all its fellow Labelers and sets its timestamp above all the numbers thus found. A Scanner reads all the Labelers, and orders their messages in accordance with their timestamps. This unbounded protocol stimulated much research, dealing mostly with the problem of devising timestamping systems with only bounded values. (See Israeli and Li 10], and Dolev and Shavit 7].) We will not deal here with this interesting problem but ask another question: to nd an unbounded timestamp algorithm in which, for any two Labelers, only one reads the other. Such an algorithm is given in Section 2 and is called \the skewed timestamp algorithm". Before answering this question, the very notion of \timestamp algorithm" must be speciied, so that the answer can be checked against the speciications. We will adapt the formal speciications of \concurrent timestamp system" (ctss) given by That the use of timestamps is old and natural is illustrated by ancient ostracons (about 800 BC) from Sumeria which show how ooerings of vin and oil were marked by date, place of origin, and destination.