... Bank of New York Mr Karl Reitz Federal Deposit Insurance Corporation, Washington, DC Mr Roger Tufts Office of the Comptroller of the Currency, Washington, DC 24 Fundamental review of ... (www.bis.org/publ/bcbs189.pdf). 14 Fundamental review of the trading book instruments that form part of a revised trading book. The Committee intends to consider the timing and scope of further work on ... scope of market risk capital requirements regardless of whether they are in the regulatory trading or banking book (with the exception of structural FX positions). 36 Fundamental review of...
... function of average cache residency time. x THE FRACTAL STRUCTUREOFDATA REFERENCE 1.16 IMS storage pools: distribution of record interarrival times. 30 1.17 TSO storage pools: distribution of ... function of 1.7 CICS storage pools: cache performance as a function of 1.8IMSstorage pools: cache performance as a function of 1.9 TSO storage pools: cache performance as a function of 1.10 ... 402.2412.341574.16.27.17.27.37.47.57.67.77.87.97.107.1 1 7.127.137.147.15Tradeoff of memory above and below the I/O interface. 6.1 Overview of free space collection results. 73 Distribution of time between track updates, for the...
... power of a memory hierarchy comes from its ability to take advantage of patterns ofdata use that are transient. Repeated requests tend to be made to a given data item over a short period of ... practical applications of such results. The focus of the applications examined in this book is on the portion of a computer system’s memory hierarchy that lies below the level of the “get” and “put” ... for dynamic migration and recall.THE FRACTAL STRUCTUREOFDATA REFERENCE Preface xviiThe primary audience intended for the book is the large body of workers who endeavor to ensure that computer...
... definition of Ω(l) to obtain the following upper bound on the hit ratio h:THE FRACTAL STRUCTUREOFDATA REFERENCE (1.1)where lc≈ .0008 is the fraction of disk tracks capable of being stored ... a given data item is not fixed with time. Instead, it depends strongly upon the amount of time that has passed since the previous reference. In a study of any large collection ofdata items ... statement Ω(.1) = .5 wouldmean that fifty percent of the total requests are to data contained in the busiest ten percent of the tracks. Since the beginning of the sorted list represents the best possible...
... to the data. If we now examine the structureof database software, in an effort to account for data reuse at a variety of time scales, we find that we need not look far. For example, data reuse ... application data, but also a few of the curves for system data. THE FRACTAL STRUCTUREOFDATA REFERENCE 2. Slopes between 0.3 and 0.4. This group consists almost entirely of system data Suppose, ... 6THE FRACTAL STRUCTUREOFDATA REFERENCE 3. MODEL DEFINITION Eventually, this book will present abundant statistical summaries of data reference patterns. As a starting...
... FRACTAL STRUCTUREOFDATA REFERENCE Therefore,By contrast, measurements of the single-reference residency time τ requirespecific instrumentation that is not available as a part of standard ... examine closely the structureof such visits, which the hierarchical reuse model predicts to be predominately transient. Based upon our analysis of 14for the averages of these three quantities. ... example, the minidisk cache facility mentioned earlier manages data in units of one page frame (oneblock of 4096 bytes). The unit ofgranularity used in managing the cache has an important effectupon...
... 20 THE FRACTAL STRUCTUREOFDATA REFERENCE Figure 1.7. time.CICS storage pools: cache performance as a function of average cache residency Figure 1.8. time.IMS ... function of cache size. Just as the miss ratio, as a function of the single-reference residency time, takes the form of a simple power law, so does the miss ratio, as a function of cache ... (1.23)The existence of a power law relationship between these quantities was first noted by Chow [ 15], and demonstrated convincingly by Smith [ 16]. 16 THE FRACTAL STRUCTUREOFDATA REFERENCE But...
... At this time, the use of explicit controls to restrict the use of cache memory remained common. Today, storage controls offer far larger amounts of memory. Most models of storage control, for ... such data is typically freed long before the data would have progressed from the top to the bottom of the LRU list. Storage controls also typically use sequential prestage operations to bring data ... given item of sequential data tends to be either immediate, or a relatively long time in the future. As a result, most cached storage controls perform early demotion of sequential data residing...
... amount of time t has passed since some given I/O request. The behavior of λ(t) providesan alternative method of characterizing the hierarchical reuse model, which 34THE FRACTAL STRUCTUREOFDATA ... the 3380 family of storage devices. 2 Assumes a track belonging to the 3390 family of storage devices. Hierarchical Reuse Model 3 1 Figure 1.17. TSO storage pools: distribution of track interarrival ... reuse model. This results in a series of equations of the form (1.5), one for each workload. In graphical terms, it corresponds to fitting each workload’s plot of interarrival statistics with a...
... thinking of the available tracks as being the leaves of a binary tree, of height Hmax. The ancestors of any given track number l are identified implicitly by the binary representation of the ... values of the interarrival time, or it may reflect a second-order effect not captured by the approximate method of analysis adopted in the previous section. THE FRACTAL STRUCTUREOFDATA REFERENCE ... THE FRACTAL STRUCTUREOFDATA REFERENCE (2-3)Given this description of F(.), we may apply the David-Johnson extremevalue approximation [20] to estimate the smallest value γhi of the random...
... variety of individual “daemons”, large numbers of which can run concurrently and independently. This type of benchmark structure is attractive, in that it mirrors, at a high level, the behavior of ... concept of stackdistance, or the depth at which previously referenced data items appear in the LRU list [8, 21]. The idea is to build a history of previous references (organized in the form of ... well.42 THE FRACTAL STRUCTUREOFDATA REFERENCE A general-purpose technique exists for generating synthetic patterns of ref-erence, which is also capable of producing references that conform...
... applications, of 72 percent. THE FRACTAL STRUCTUREOFDATA REFERENCE As a final step, we must also consider how to round off the computed cache memory requirement of 2256 megabytes. Since this requirement ... control of this type, virtually all write requests will typically be reported as “hits,” even though some of them may require allocation of memory. For database I/O, this potential source of error ... .2. Then the absolute value of ζiis no greater than .1/.7 = .14 for either workload. As a result, the absolute value of either of the second order summation terms of (3.11), calculated without...