Dynamic Verification of Memory Consistency in Cache-Coherent Multithreaded Computer Architectures

Multithreaded servers with cache-coherent shared memory are the dominant type of machines used to run critical network services and database management systems. To achieve the high availability required for these tasks, it is necessary to incorporate mechanisms for error detection and recovery. Correct operation of the memory system is defined by the memory consistency model. Errors can therefore be detected by checking if the observed memory system behavior deviates from the specified consistency model. Based on recent work, we design a framework for dynamic verification of memory consistency (DVMC). The framework consists of mechanisms to verify three invariants that are proven to guarantee that a specified memory consistency model is obeyed. We describe an implementation of the framework for the SPARCv9 architecture, and we experimentally evaluate its performance using full-system simulation of commercial workloads.     

Effect of brain edema on the recurrence pattern of malignant gliomas

PURPOSE: To assess the influence of initial preoperative brain edema in malignant gliomas on regrowth patterns. SUBJECTS AND METHODS: 79 patients with histologically verified supratentorial malignant glioma were prospectively studied by magnetic resonance imaging (MRI) before and every 2-3 months after surgery. The median follow-up time was 11 months. We correlated the configuration of the initial vasogenic edema on T2-weighted images with tumor regrowth patterns on contrast-enhanced T1-weighted images. RESULTS: 35/47 tumor regrowths (75%) imitated the initial edema configuration, while 11/47 occurred within the initial tumor bed; in one case tumor recurrence was multilocal. CONCLUSION: In glioblastoma, tumor regrowth patterns correlate positively with the configuration of the initial vasogenic brain edema. The initial, "presurgical" peritumoral edema should thus be considered when planning further treatment.     

Staphylococcal enterotoxin B primes cytokine secretion and lytic activity in response to native bacterial antigens

Superantigens stimulate T-lymphocyte proliferation and cytokine production, but the effects of superantigen exposure on cell function within a complex, highly regulated immune response remain to be determined. In this study, we demonstrate that superantigen exposure significantly alters the murine host response to bacterial antigens in an in vitro coculture system. Two days after exposure to the superantigen staphylococcal enterotoxin B, splenocytes cultured with Streptococcus mutans produced significantly greater amounts of gamma interferon (IFN-gamma) and interleukin-12 than did sham-injected controls. The majority of IFN-gamma production appeared to be CD8(+) T-cell derived since depletion of this cell type dramatically reduced the levels of IFN-gamma. To study host cell damage that may occur following superantigen exposure, we analyzed cytotoxicity to "bystander" fibroblast cells cultured with splenocytes in the presence of bacterial antigens. Prior host exposure to staphylococcal enterotoxin B significantly enhanced fibroblast cytotoxicity in the presence of bacteria. Neutralization of IFN-gamma decreased the amount of cytotoxicity observed. However, a greater reduction was evident when splenocyte-bacterium cocultures were separated from the bystander cell monolayer via a permeable membrane support. Increased cytotoxicity appears to be primarily dependent upon cell-cell contact. Collectively, these data indicate that overproduction of inflammatory cytokines may alter the activity of cytotoxic immune cells. Superantigen exposure exacerbates cytokine production and lytic cell activity when immune cells encounter bacteria in vitro and comparable activities could possibly occur in vivo.     

Calcium can disrupt the SNARE protein complex on sea urchin egg secretory vesicles without irreversibly blocking fusion

The homotypic fusion of sea urchin egg cortical vesicles (CV) is a system in which to correlate the biochemistry and physiology of membrane fusion. Homologues of vesicle-associated membrane protein (VAMP), syntaxin, and SNAP-25 were identified in CV membranes. A VAMP and syntaxin immunoreactive band at a higher apparent molecular mass (approximately 70 kDa) was detected; extraction and analysis confirmed that the band contained VAMP, SNAP-25, and syntaxin. This complex was also identified by immunoprecipitation and by sucrose gradient analysis. VAMP in the complex was insensitive to proteolysis by tetanus toxin. All criteria identify the SNARE complex as that described in other secretory systems. Complexes exist pre-formed on individual CV membranes and form between contacting CV. Most notably, CV SNARE complexes are disrupted in response to [Ca2+]free that trigger maximal fusion. N-Ethylmaleimide, which blocks fusion at or before the Ca2+-triggering step, blocks complex disruption by Ca2+. However, disruption is not blocked by lysophosphatidylcholine, which transiently arrests a late stage of fusion. Since removal of lysophosphatidylcholine from Ca2+-treated CV is known to allow fusion, complex disruption occurs independently from the membrane fusion step. As Ca2+ disrupts rather than stabilizes the complex, the presumably coiled-coil SNARE interactions are not needed at the time of fusion. These findings rule out models of fusion in which SNARE complex formation goes to completion ("zippers-up") after Ca2+ binding removes a "fusion-clamp."     

A SIMD-efficient 14 instruction shader program for high-throughput microtriangle rasterization

This paper shows that breaking the barrier of 1 triangle/clock rasterization rate for microtriangles in modern GPU architectures in an efficient way is possible. The fixed throughput of the special purpose culling and triangle setup stages of the classic pipeline limits the GPU scalability to rasterize many triangles in parallel when these cover very few pixels. In contrast, the shader core counts and increasing GFLOPs in modern GPUs clearly suggests parallelizing this computation entirely across multiple shader threads, making use of the powerful wide-ALU instructions. In this paper, we present a very efficient SIMD-like rasterization code targeted at very small triangles that scales very well with the number of shader cores and has higher performance than traditional edge equation based algorithms. We have extended the ATTILA GPU shader ISA (del Barrioet al. in IEEE International Symposium on Performance Analysis of Systems and Software, pp. 231–241, 2006) with two fixed point instructions to meet the rasterization precision requirement. This paper also introduces a novel subpixel Bounding Box size optimization that adjusts the bounds much more finely, which is critical for small triangles, and doubles the 2×2-pixel stamp test efficiency. The proposed shader rasterization program can run on top of the original pixel shader program in such a way that selected fragments are rasterized, attribute interpolated and pixel shaded in the same pass. Our results show that our technique yields better performance than a classic rasterizer at 8 or more shader cores, with speedups as high as 4× for 16 shader cores.     

3-D B-spline Wavelet-Based Local Standard Deviation (BWLSD): Its Application to Edge Detection and Vascular Segmentation in Magnetic Resonance Angiography

Extracting reliable image edge information is crucial for active contour models as well as vascular segmentation in magnetic resonance angiography (MRA). However, conventional edge detection techniques, such as gradient-based methods and wavelet-based methods, are incapable of returning reliable detection responses from low contrast edges in the images. In this paper, we propose a novel edge detection method by combining B-spline wavelet magnitude with standard deviation inside local region. It is proved theoretically and demonstrated experimentally in this paper that the new edge detection method, namely BWLSD, is able to give consistent and reliable strengths for edges with different image contrasts. Moreover, the relationship between the size of local region with non-zero wavelet magnitudes and the scale of wavelet function is established. This relationship indicates that if the scale of the adopted wavelet function is s, then the size of a local region, from which the standard deviation is estimated, should be 2s−1. The proposed edge detection technique is embedded in FLUX, namely, BWLSD-FLUX, for vascular segmentation in MRA image volumes. Experimental results on clinical images show that, as compared with the conventional FLUX, BWLSD-FLUX can achieve better segmentations of vasculatures in MRA images under same initial conditions.     

Anisotropic Smoothness Classes: From Finite Element Approximation to Image Models

We propose and study quantitative measures of smoothness f ? A(f) which are adapted to anisotropic features such as edges in images or shocks in PDE’s. These quantities govern the rate of approximation by adaptive finite elements, when no constraint is imposed on the aspect ratio of the triangles, the simplest example being \(A_{p}(f)=\|\sqrt{|\mathrm{det}(d^{2}f)|}\|_{L^{\tau}}\) which appears when approximating in the L ^p norm by piecewise linear elements when \(\frac{1}{\tau}=\frac{1}{p}+1\). The quantities A(f) are not semi-norms, and therefore cannot be used to define linear function spaces. We show that these quantities can be well defined by mollification when f has jump discontinuities along piecewise smooth curves. This motivates for using them in image processing as an alternative to the frequently used total variation semi-norm which does not account for the smoothness of the edges.     

Approximated approximations for SAO

We propose to replace a number of popular approximations by their diagonal quadratic Taylor series expansions. The resulting separable quadratic approximations are easily convexified, and are well suited for use in dual sequential approximate optimization (SAO) algorithms. Global convergence of the resulting SAO algorithms may be enforced in a natural way using conservatism. The approximated approximation approach is explicitly illustrated for (i) reciprocal and exponential intervening variables, (ii) the intervening variables used in the method of moving asymptotes (MMA), (iii) the intervening variables used in CONLIN, and (iv) the TANA-3 approximations. The use of intermediate responses for use in, for example, truss and frame-like structures, is also discussed. Key advantages of replacing nonlinear approximations by their diagonal quadratic approximations are that these approximated approximations can all be used simultaneously in a single dual statement; the dual does not depend on the form of the original approximations. In addition, in a dual setting, the resulting subproblems yield simple analytical relationships between the primal and dual variables, which is often not the case with the original nonlinear approximations. An important example hereof is the exponential approximation. Although the diagonal quadratic approximations may differ notably from their original counterparts, they typically are quite similar in a sufficiently small search subregion, which relates to the move limits commonly used in SAO anyway.     

IEEE 802.21: Media independence beyond handover

The IEEE 802.21 standard facilitates media independent handovers by providing higher layer mobility management functions with common service primitives for all technologies. Right after the base specification was published, several voices rose up in the working group advocating to broaden the scope of IEEE 802.21 beyond handovers. This paper aims at updating the reader with the main challenges and functionalities required to create a Media Independence Service Layer, through the analysis of scenarios which are being discussed within the working group: 1) Wireless Coexistence, and 2) Heterogeneous Wireless Multihop Backhaul Networks.