Balanced computer systemsThe central server model is used to extend Buzen's results on balance and bottlenecks. Two measures are developed which appear to be useful for evaluating and improving computer system performance. The first measure, called the balance index, is useful for balancing requests to the peripheral processors. The second quantity, called the sensitivity index, indicates which processing rates have the most effect on overall system performance. |
Contents
Balancing Requests to Peripheral Processors | 4 |
Designing a Balanced System | 13 |
Computation of Balance Sensitivity and Capacity | 22 |
2 other sections not shown
Common terms and phrases
achieve capacity Air Force Andrews AFB Arlington Army Electronics Command average waiting b₁ balancing requests bottleneck Brooks AFB Buzen California central server model Command Fort Monmouth Commander Naval condition maximizes conditions for balance conditions given consider the problem constraint convex function CPU bound CPU utilization decrease degree of multiprogramming devices Director discernible interval entropy equivalent Euler's theorem example with probabilities f a+b f(a+b geometric mean given by 9 Hanscom Field Bedford highest utilization homogeneous functions I/O bound improve performance incremental amount indexes are equal job mix JSEP L. G. Hanscom Field L₁ Lemma Massachusetts 01730 maximizes our uncertainty maximum throughput measure Mexico 88002 Attn n-1 a+b Naval Research Laboratory Office of Naval P₁ peripheral processor probabilities probability vectors problem of balancing Proof Sands Missile Range smaller balance index symmetric function Technical Library U.S. Army Electronics Virginia White Sands Missile Wright-Patterson AFB эт