Budy D. Notohardjono

High-end server low-temperature cooling

The IBM S/390® G4 CMOS system, first shipped in 1997, was the first high-end system to use refrigeration. The decision to employ refrigeration cooling instead of other cooling options such as high-flow air cooling or various water-cooling schemes focused on the potential system performance improvement obtainable by lowering coolant temperatures using a refrigeration system. This paper reviews the historical background of refrigeration from its use in the early 1800s to its implementation in computer systems in the early 1990s. The advantages and disadvantages of using refrigeration in the cooling of computer systems are examined. The advantages have outweighed the disadvantages, leading to the first use by IBM of refrigeration in cooling the S/390 G4 server. The design of the refrigeration system for the S/390 G4 system is described in detail, and some of the key parametric studies that contributed to the final design are described.

A power, packaging, and cooling overview of the IBM eServer z900

This paper provides an overview of the power, packaging, and cooling aspects of the IBM eServer z900 design. The semiconductor processor chips must be supported and protected in a mechanical structure that has to provide electrical interconnects while maintaining the chip junction temperature within specified limits. The mechanical structure should be able to withstand shock and vibrations during transportation or events such as earthquakes. The processor chips require electrical power at well-regulated voltages, unaffected by the ac-line voltage and load current fluctuations. The acoustical and electromagnetic noise produced by the hardware must be within the limits set by national regulatory agencies, and the electronic operations must be adequately protected from disruption caused by electromagnetic radiation. For high availability, the power, packaging, and cooling hardware must have redundancy and the ability to be maintained while the system is operating. This paper first overviews the packaging hardware, followed by a description of the first- and second-level packaging, which includes the mother board and the multichip module. Thermal management is discussed from the point of view of both the multichip module and the overall system. Power conversion, management, and distribution are presented next. Finally, the design aspects involved with meeting the requirements of electromagnetic compatibility, acoustics, and immunity to shock, vibration, and earthquakes are discussed.

Modular server frame with robust earthquake retention

Adequate retention of computer systems during earthquake events is important because it can not only prevent human injury and potential system damage, but also ensure system availability by limiting to acceptable levels the transmitted accelerations to critical system components such as hard drives. This paper discusses the design of an IBM frame structure and related hardware, and the retention methods used, to provide a robust mechanical installation in both raised- and nonraised-floor environments, capable of surviving severe seismic events. The development of the frame structure and the retention hardware involves extensive earthquake simulation testing, in which the responses of the system under different earthquake test profiles are recorded and analyzed in both the time and frequency domains. Industry standards such as the Bellcore NEBS GR-63-CORE and IBM internal specifications are reviewed and compared, and the transient responses of competing frame designs subjected to various earthquake profiles are investigated to ensure compliance. Finally, the concept of modular design, in which various frame components are utilized to create a flexible family of frames, is discussed.

Static and Dynamic Handling Stability of Server Rack Computers

This paper discusses the static and dynamic stability analysis of rack or frame computer/server products during shipping and relocation. The static stability is the ability of server products to resist tipping over on a typical raised floor in a datacenter or when it is installed in its operational product environment. The dynamic stability is the ability to resist tipping over when a velocity change occurs during re-location either on flat or inclined planes. The product consists of a frame or a rack in which components such as processor units, input-output units and power supplies are installed.The static stability analysis presented here calculates the tip over threshold angle, which is the maximum angle of an inclined plane on which the product can be placed without tipping over. The location of the installed components in a frame, the dimension and weight of the installed components, and the dimension of the product dictate the overall static stability of the product. Specifically, those parameters affect the location of the center of gravity of the product and the tip over threshold angle. The tip over threshold angle is a critical parameter influencing the dynamic stability of the product..The dynamic stability of an unpackaged product moving on casters can be calculated using the conservation of mechanical energy principle. Finite element modeling is a good way to evaluate the dynamic stability of a product during manual handling or mechanical handling; for instance, on a forklift. The objective of the finite element modeling is to provide guidelines on the maximum speed, minimum radius curvature, and safe turning speed of a forklift when transporting a product.The main objective of the analysis presented here is to provide a method for analyzing the static and dynamic stability of a rack style computer server product during shipping, relocation, and handling.Copyright

Modeling of a Mainframe Server Frame Subjected to Seismic Loadings

This paper describes the finite element modeling of a mainframe server frame. The frame consists of a server rack or frame with its add-on stiffening brackets. The frame is anchored directly to the floor with bolts at each of the four corners. The Telcordia Zone 4 earthquake test profile represents a severe dynamic load input and will be used herein to analyze the mainframe server frame. The main objective of this modeling is to validate the frame design prior to actual seismic testing, which ultimately ensures the structural integrity of a functional mainframe system during a seismic event. The server frame finite element (FE) model is derived from a three dimensional CAD model of a standard sheet metal frame weldment assembly which is then simplified and meshed with finite elements. This FE model represents the server frame, welded connections, and stiffening brackets, which are specifically designed to withstand seismic test profiles. To represent the components that populate the server frame, point masses are tied to the frame at the same attachment points that exist in the real assembly. The validation of the FE model involves the use of a horizontal shaker test to assess the server frame’s stiffness. The goal of this paper is to show a good correlation between FE model and test results using two separate FE solver technologies: implicit and explicit. For an implicit solver, linear material properties were used to obtain modal behavior that approximates the actual server frame’s behavior. Once these outputs were achieved, further response refinement was attempted by porting the model to an explicit dynamic solver. An explicit solver allowed non-linear material properties and body to body contact behavior to be included in the FE model while applying the seismic test profile to the server frame using a time domain input. The explicit dynamic model outputs used to correlate to actual test results were the modal dynamics, the displacement of the top of the server frame, and the maximum reaction force at the anchored corners. Finally, a functional system was subjected to the Telcordia Zone 4 seismic test profile. The system was functional during and after the seismic test with no significant structural damage having occurred.Copyright

SEISMIC TESTING AND ANALYSIS OF MAIN FRAME COMPUTER STRUCTURE

Designing a mainframe computer structure that can withstand seismic events requires significant testing and analysis. The computer frame and anchorage system must have adequate strength and stiffness to counteract earthquake-induced forces, thereby preventing human injury and potential system damage. However, this same frame and anchorage system must also meet the requirement of ensuring continued system operation by limiting overall displacement of the structure to accepted levels. Therefore, the test and analysis scope need to include the mainframe structure and its anchorage attachment to the building’s concrete floor via a raised floor structure. This paper discusses the numerical modeling and its verification to quantify the robustness of a high end computer server structure subjected to a severe seismic event. The frame of the computer is the structure where components are installed (e.g., the central processing unit, the input-output drawer, the power supply component, etc.). The dynamic response of this structure is highly related to the weight of the components, the assembly’s inherent natural frequency, and the location of the structure’s center of gravity. The natural frequency of various mainframe configurations were analyzed and measured by either changing the weights (i.e., adding or eliminating components) or by changing the structural stiffness (e.g., adding reinforcement brackets). The main objective of the modeling was to ensure structural integrity following a seismic test of a functional server system. Finite element analysis (FEA) was employed as part of the overall frame’s structural robustness design verification, whereby the simulated modal analysis results were compared to both the experimental modal data of the frame structure as well as measured swept sine data. This design study builds toward the objective of constructing a verified model of the server frame and components, which lead to a guideline for implementing optimized reinforcement. As part of the verification, the mainframe structures were subjected to horizontal table vibration tests simulating the loads and environmental conditions endured during seismic events. During experimental verification, the dynamic responses were recorded and analyzed in both the time and frequency domains.

Seismic Evaluation of Large Server Computer Structure

This paper discusses the analysis and verification of a finite element model which simulates the robustness of a high end computer server structure during a severe seismic event. The server consists of the frame which is the structure that components are installed into, such as processor units, input-output units and power components. The finite element modeling of this server frame is presented here not only to inform on creating an accurate model for simulation purposes, but also to provide guidelines as to the critical factors in setting up a large assembly finite element model (FEM) and to establish the optimum methodology for modeling this complex assembly with the available analysis software tools. For verification, the simulated modal data is compared to both modal data measured from an instrumented impact hammer, and to measured swept sine data. The simulated results compare favorably with the measured data, and it has been determined that location and integrity of the welded connections are critical for an accurate vibration response of the finite element model. The analysis frame model was subjected to loads and environmental conditions similar to those endured under horizontal table vibration tests and seismic events. The results of the experimental testing and simulations were compared and proved to be in a good correlation. Based on this verified finite element model, any additional redesign of the frame structure and its stiffening members can proceed very efficiently. This design study builds toward the objective of constructing a verified model of the server frame and components which will lead to a guideline for implementing stiffener designs on high-end server systems.Copyright

Structural Analysis of High-End Server Computer Frames Under Earthquake Loading

This paper reports the mechanical design, structural analysis, and experimental correlation of bracing concepts for high-end computer servers subjected to loads simulating earthquake conditions. The development and evaluation of these stiffening alternatives follows a step-by-step process of finite element analysis coupled with parallel experimental testing. The numerical model is derived from the simplified CAD geometry of an existing server frame. An analysis of this frame model is subjected to a load environment similar to those endured under actual horizontal table vibration tests. The result of this series of analyses is a design study examining how a range of bracing designs affects the global frame rigidity. This design study builds toward the objective of constructing a verified model of the server frame and components that will lead to a guideline for implementing stiffener designs on high-end server systems.Copyright