The Cell Broadband Engine: Exploiting Multiple Levels of Parallelism in a Chip Multiprocessor

As CMOS feature sizes continue to shrink and traditional microarchitectural methods for delivering high performance (e.g., deep pipelining) become too expensive and power-hungry, chip multiprocessors (CMPs) become an exciting new direction by which system designers can deliver increased performance. Exploiting parallelism in such designs is the key to high performance, and we find that parallelism must be exploited at multiple levels of the system: the thread-level parallelism that has become popular in many designs fails to exploit all the levels of available parallelism in many workloads for CMP systems. We describe the Cell Broadband Engine and the multiple levels at which its architecture exploits parallelism: data-level, instruction-level, thread-level, memory-level, and compute-transfer parallelism. By taking advantage of opportunities at all levels of the system, this CMP revolutionizes parallel architectures to deliver previously unattained levels of single chip performance. We describe how the heterogeneous cores allow to achieve this performance by parallelizing and offloading computation intensive application code onto the Synergistic Processor Element (SPE) cores using a heterogeneous thread model with SPEs. We also give an example of scheduling code to be memory latency tolerant using software pipelining techniques in the SPE. This paper is based in part on “Chip multiprocessing and the Cell Broadband Engine”, ACM Computing Frontiers 2006.     

Dynamic and transparent binary translation

High-frequency design and instruction-level parallelism (ILP) are important for high-performance microprocessor implementations. The Binary-translation Optimized Architecture (BOA), an implementation of the IBM PowerPC family, combines binary translation with dynamic optimization. The authors use these techniques to simplify the hardware by bridging a semantic gap between the PowerPC's reduced instruction set and even simpler hardware primitives. Processors like the Pentium Pro and Power4 have tried to achieve high frequency and ILP by implementing a cracking scheme in hardware: an instruction decoder in the pipeline generates multiple micro-operations that can then be scheduled out of order. BOA relies on an alternative software approach to decompose complex operations and to generate schedules, and thus offers significant advantages over purely static compilation approaches. This article explains BOA's translation strategy, detailing system issues and architecture implementation     

TRANSPORT PHENOMENA IN MICRO HEAT EXCHANGERS WITH CORRUGATED WALLS

Heat exchangers with corrugated walls are well known as macro heat exchangers. In previous systematic analyses heat exchangers with crosswise corrugated walls at large Reynolds numbers were analyzed. The wall corrugation was generated by sinusoidally shaped walls as well as by crosswise corrugated cylinders of different cross sections on plane walls, such as those being used as spacers in membrane technology. The low-Reynolds-number range of these structures is also interesting for application as micro heat exchangers. Therefore first results will be presented, where such structures are experimentally analyzed in the low-Reynolds-number range up to Re S 10, especially interesting for applications as micro heat exchangers. For the use of corrugated walls in the free-flow arrangement, first new results at low Reynolds numbers for this heat exchanger geometry are presented, where the transition from turbulent to laminar flow range is dominated by the Görtler-Dean instability.     

Josephson junctions with 1 ?m dimensions and with picosecond switching times

Josephson junctions with 1 × 3 ?m dimensions have no resonances. For a current density of about 1000 A/cm2, the risetime of the switching transient from zero to the gap voltage is below 38 ps.     

Novel glycoinositolphosphosphingolipids, basidiolipids, from Agaricus

From the edible mushroom, the basidiomycetes Agaricus bisporus and Agaricus campestris, a novel carbohydrate-homologous series of four glyco-inositol-phospho-sphingolipids, designated basidiolipids, was isolated and the constituents purified. The chemical structures of the basidiolipids were elucidated to be: Manpbeta1-2inositol1-phospho-ceramide, Galpalpha-6[Fucpalpha-2]Galpbeta-6Manpbeta-2i nositol1-phospho-ceramide, Galpalpha-6Galpalpha-6[Fucpalpha-2]Galpbeta- 6Manpbeta-2inositol1-phospho-ceramide and Galpalpha-6Galpalpha-6Galpalpha-6[Fucpalpha-2] Galpbeta-6Manpbeta-2ino sitol1-phospho-ceramide. All four glycolipids contained a ceramide which was composed of phytosphingosine and predominantly alpha-hydroxy-behenic and alpha-hydroxy-lignoceric acid.     

Insights Into the Micelle-Induced β-Hairpin-to-α-Helix Transition of a LytA-Derived Peptide by Photo-CIDNP Spectroscopy

A choline-binding module from pneumococcal LytA autolysin, LytA_239–252, was reported to have a highly stable nativelike β-hairpin in aqueous solution, which turns into a stable amphipathic α-helix in the presence of micelles. Here, we aim to obtain insights into this DPC-micelle triggered β-hairpin-to-α-helix conformational transition using photo-CIDNP NMR experiments. Our results illustrate the dependency between photo-CIDNP phenomena and the light intensity in the sample volume, showing that the use of smaller-diameter (2.5 mm) NMR tubes instead of the conventional 5 mm ones enables more efficient illumination for our laser-diode light setup. Photo-CIDNP experiments reveal different solvent accessibility for the two tyrosine residues, Y249 and Y250, the latter being less accessible to the solvent. The cross-polarization effects of these two tyrosine residues of LytA_239–252 allow for deeper insights and evidence their different behavior, showing that the Y250 aromatic side chain is involved in a stronger interaction with DPC micelles than Y249 is. These results can be interpreted in terms of the DPC micelle disrupting the aromatic stacking between W241 and Y250 present in the nativelike β-hairpin, hence initiating conversion towards the α-helix structure. Our photo-CIDNP methodology represents a powerful tool for observing residue-level information in switch peptides that is difficult to obtain by other spectroscopic techniques.