Soft Error Resiliency (SER) is a major concern for Petascale high performance computing (HPC) systems. In designing Blue Gene/Q (BG/Q) [8], many mechanisms were deployed to target SER including extensive use of Silicon-On-Insulator (SOI), radiation-hardened latches [7,13], detection and correction in on-chip arrays, and very low radiation packaging materials. On the other hand, it is well known that application behavior has major impacts on the masking (or “derating” factor) in system SER calculations. The principal goal of this project is to understand the interaction between BG/Q hardware and high-performance applications when it comes to SER by performing and evaluating a chip irradiation experiment.

Fault injection through accelerated irradiation is an effective way to evaluate the overall soft error resiliency of microprocessors. In this work, we report on irradiation experiments on a Blue Gene/Q (BG/Q) compute processor chip running selected applications. Blue Gene/Q is the third generation of IBM's massively parallel, energy efficient Blue Gene series of supercomputers. In the experiments, we found 69 code fails. Out of these, 26 code fails are relevant for the calculation of the mean-time-between-failures (MTBF) for a 20 PetaFLOP, 96 rack system running a comparable workload mix. The expected MTBF for check-stops due to cosmic radiation and alpha particles from chip packaging materials is calculated to be 51 days for sea-level at New York City running the application mix studied. If the most vulnerable application is run exclusively, the projected MTBF is 35 days. These are outstanding results for a machine of this magnitude. The beaming experiment and projected MTBF validate the necessity to include autonomous hardware detection and recovery at the cost of design effort, silicon area and power.

A 32nm SOI critical path monitor (CPM) that can provide timing measurements to a Digital PLL for dynamic frequency adjustments in the 8-core POWER7+™ microprocessor is described. The CPM calibrates to within 2% of cycle time from nominal to turbo voltages. Its voltage sensitivity is 10mV/bit. It tracks processor temperature sensitivity to within 1.5% of nominal frequency, and has a sample jitter less than 1.5% of nominal frequency. The ability to detect noise dynamically allows the system to operate the processor closer to its optimal frequency for any given voltage, resulting in lower voltage for power savings or higher frequency for performance improvements.

This paper explores the performance and optimization of the IBM Blue Gene/Q (BG/Q) five dimensional torus network on up to 16K nodes. The BG/Q hardware supports multiple dynamic routing algorithms and different traffic patterns may require different algorithms to achieve best performance. Between 85% to 95% of peak network performance is achieved for all-to-all traffic, while over 85% of peak is obtained for challenging bisection pairings. A new software-controlled algorithm is developed for bisection traffic that selects which hardware algorithm to employ and achieves better performance than any individual hardware algorithm. The benefit of dynamic routing is shown for a highly non-uniform "transpose" traffic pattern. To evaluate memory and network performance, the HPCC Random Access benchmark was tuned for BG/Q and achieved 858 Giga Updates per Second (GUPS) on 16K nodes. To further accelerate message processing, the message libraries on BG/Q enable the offloading of messaging overhead onto dedicated communication threads. Several applications, including Algebraic Multigrid (AMG), exhibit from 3 to 20% gain using communication threads.

This is the first paper describing the IBM Blue Gene/Q interconnection network and message unit. The Blue Gene/Q system is the third generation in the IBM Blue Gene line of massively parallel supercomputers. The Blue Gene/Q architecture can be scaled to 20 PF/s and beyond. The network and the highly parallel message unit, which provides the functionality of a network interface, are integrated onto the same chip as the processors and cache memory, and consume 8% of the chip's area. For better application scalability and performance, we describe new routing algorithms and new techniques to parallelize the injection and reception of packets in the network interface. Measured hardware performance results are also presented.

Single-chip systems save the power dissipation that would be required for chip-to-chip communication, resulting in compact, low-power solutions for battery-powered applications. This paper describes the design and measured performance of a fully-functional digital core with a low-jitter, on-chip, MEMS-LC clock reference. This chip has been fabricated in TSMC's 0.18mum MM/RF bulk CMOS process. Maximum power consumption of the complete microsystem is 48.78mW operating at 90MHz on a 1.8V power supply

This paper describes low-power DSP architecture for use in cochlear implants. The microsystem, fabricated in TSMC 0.18mum CMOS, consumes 1.79mW from a 1.2V supply and occupies an area of 9.18mm2while providing the necessary programmability for high speech comprehension by patients. Standby power consumption is 330muW

A low-latency, HDL-synthesizable dynamic clock frequency controller is presented as a time-efficient alternative to full-custom implementations. Frequency division of a fully integrated hybrid temperature-compensated LC oscillator (TC-LCO) and ring oscillator clock reference avoids PLL locking delays to enable low-latency, hazard-free frequency selection on an actively running CPU. Fabricated in 0.18mum CMOS as part of a low-power SoC microsystem, the circuit dissipates 480muW at 1.8V

In this paper, we propose a new dual-VDD bus technique that is well suited for low power operation. This technique adapts a static pulsed bus architecture to use dual-VDD power supplies. During quiescent periods, the bus system idles at the lower of the two VDD supplies, thereby lowering static power dissipation. When actively transitioning, the inverters in the bus system are temporarily boosted to the higher VDD supply to provide the needed drive strength for performance. Since the VDD boosting is done in a pulsed manner, the bus system is in a high VDD state only when required, ensuring lower power operation without sacrificing performance. This technique yields up to a 50% reduction in total power over traditional static buses and up to a 35% reduction in total power over standard static pulsed buses, with a 12-15% delay improvement.

Modern embedded microprocessors use low power on-chip memories called scratch-pad memories to store frequently executed instructions and data. Unlike traditional caches, scratch-pad memories lack the complex tag checking and comparison logic, thereby proving to be efficient in area and power. In this work, we focus on exploiting scratch-pad memories for storing hot code segments within an application. Static placement techniques focus on placing the most frequently executed portions of programs into the scratch-pad. However, static schemes are inherently limited by not allowing the contents of the scratch-pad memory to change at run time. In a large fraction of applications, the instruction memory footprints exceed the scratch-pad memory size, thereby limiting the usefulness of the scratch-pad. We propose a compiler managed dynamic placement algorithm, wherein multiple hot code sequences, or traces, are overlapped with each other in the scratch-pad memory at different points in time during execution. Special copy instructions are provided to copy the traces into the scratch-pad memory at run-time. Using a power estimate, the compiler initially selects the most frequent traces in an application for relocation into the scratch-pad memory. Through iterative code motion and redundancy elimination, copy instructions are inserted in infrequently executed regions of the code. For a 64-byte code cache, the compiler managed dynamic placement achieves an average of 64% energy improvement over the static solution in a low-power embedded microcontroller.

Low-power embedded processors utilize compact instruction encodings to achieve small code size. Such encodings place tight restrictions on the number of bits available to encode operand specifiers and, thus, on the number of architected registers. As a result, performance and power are often sacrificed as the burden of operand supply is shifted from the register file to the memory due to the limited number of registers. In this paper, we investigate the use of a windowed register file to address this problem by providing more registers than allowed in the encoding. The registers are organized as a set of identical register windows where, at each point in the execution, there is a single active window. Special window management instructions are used to change the active window and to transfer values between windows. This design gives the appearance of a large register file without compromising the instruction encoding. To support the windowed register file, we designed and implemented a graph partitioning-based compiler algorithm that partitions program variables and temporaries referenced within a procedure across multiple windows. On a 16-bit embedded processor, an average of 11 percent improvement in application performance and 25 percent reduction in system power was achieved as an 8-register design was scaled from one to two windows.

Low-power embedded processors utilize compact instruction encodings to achieve small code size. Instruction sizes of 8 to 16 bits are common. Such encodings place tight restrictions on the number of bits available to encode operand specifiers, and thus on the number of architected registers. The central problem with this approach is that performance and power are often sacrificed as the burden of operand supply is shifted from the register file to the memory due to the limited number of registers. In this paper, we investigate the use of a windowed register file to address this problem by providing more registers than allowed in the encoding. The registers are organized as a set of identical register windows where at each point in the execution there is a single active window. Special window management instructions are used to change the active window and to transfer values between windows. The goal of this design is to give the appearance of a large register file without compromising the instruction encoding. To support the windowed register file, we designed and implemented a novel graph partitioning based compiler algorithm that partitions virtual registers within a given procedure across multiple windows. On a 16-bit embedded processor with a parameterized register window, an average of 10% improvement in application performance and 7% reduction in system power was achieved as an eight-register design was scaled from one to four windows.

In this work, we report on an unprecedented design where digital, analog, and MEMS technologies are combined to realize a general-purpose single-chip CMOS microsystem. The convergence of these technologies has enabled the development of a low power, portable microinstrument ideally suited for controlling environmental and bio-implantable sensors.

An overview of microsystems technology is presented along with a discussion of the recent trends and challenges associated with its development. A typical bottom-up design methodology is described and we propose, in contrast, an efficient and effective top-down methodology. We illustrate its implementation with the development of a microsystem design that has been completed and fabricated in CMOS technology. Gaps in the tool capabilities are identified and suggestions for future directions in CAD tool support for microsystems technology are presented.