Efficient application identification and the temporal and spatial stability of classification schema

Motivated by the importance of accurate identification for a range of applications, this paper compares and contrasts the effective and efficient classification of network-based applications using behavioral observations of network-traffic and those using deep-packet inspection.Importantly, throughout our work we are able to make comparison with data possessing an accurate, independently determined ground-truth that describes the actual applications causing the network-traffic observed.In a unique study in both the spatial-domain: comparing across different network-locations and in the temporal-domain: comparing across a number of years of data, we illustrate the decay in classification accuracy across a range of application–classification mechanisms. Further, we document the accuracy of spatial classification without training data possessing spatial diversity.Finally, we illustrate the classification of UDP traffic. We use the same classification approach for both stateful flows (TCP) and stateless flows based upon UDP. Importantly, we demonstrate high levels of accuracy: greater than 92% for the worst circumstance regardless of the application.     

Transparent autonomization in CORBA

Increasingly, software systems are constructed by integrating and composing multiple existing applications. The resulting complexity increases the need for self-management of the system. However, adding autonomic behavior to composite systems is difficult, especially when the constituent components are heterogeneous and they were not originally designed to support such interactions. Moreover, entangling the code for self-management with the code for the business logic of the original applications may actually increase the complexity of the systems, counter to the desired goal. In this paper, we address autonomization of composite systems that use CORBA, one of the first widely used middleware platforms introduced more than 17 years ago that is still commonly used in numerous systems. We propose a model, called Adaptive CORBA Template (ACT), that enables autonomic behavior to be added to CORBA applications automatically and transparently, that is, without requiring any modifications to the code implementing the business logic of the original applications. To do so, ACT uses “generic” interceptors, which are added to CORBA applications at startup time and enable autonomic behavior to be introduced later at runtime. We have developed ACT/J, a prototype of ACT in Java. We describe a case study in which ACT/J is used to introduce three types of autonomic behavior (self-healing, self-optimization, and self-configuration) to a distributed surveillance application.     

Mixed-reality environment for frictional parameters identification in servo-pneumatic system

This paper outlines a method to identify the friction parameters for servo-pneumatic systems using a mixed-reality environment. To acquire system friction parameters accurately can be extremely difficult once the servo-system has been assembled because of its highly nonlinear nature, which causes a great difficulty in servo-pneumatic system modelling and control. In this research, a mixed-reality environment has been employed to determine the friction parameters effectively and efficiently through online identification. Traditionally, friction parameters identification can be performed manually or automatically using traditional optimization methods or modern ones such as neural networks. The advantages of the proposed method are the high accuracy in the estimated parameters, its simplicity and its speed. An experimental case study has been conducted and the results showed the accuracy and effectiveness of the proposed method.     

Optimization of dynamic response using a monolithic-time formulation

The design of systems for dynamic response may involve constraints that need to be satisfied over an entire time interval or objective functions evaluated over the interval. Efficiently performing the constrained optimization is challenging, since the typical response is implicitly linked to the design variables through a numerical integration of the governing differential equations. Evaluating constraints is costly, as is the determination of sensitivities to variations in the design variables. In this paper, we investigate the application of a temporal spectral element method to the optimization of transient and time-periodic responses of fundamental engineering systems. Through the spectral discretization, the response is computed globally, thereby enabling a more explicit connection between the response and design variables and facilitating the efficient computation of response sensitivities. Furthermore, the response is captured in a higher order manner to increase analysis accuracy. Two applications of the coupling of dynamic response optimization with the temporal spectral element method are demonstrated. The first application, a one-degree-of-freedom, linear, impact absorber, is selected from the auto industry, and tests the ability of the method to treat transient constraints over a large-time interval. The second application, a related mass-spring-damper system, shows how the method can be used to obtain work and amplitude optimal time-periodic control force subject to constraints over a periodic time interval. This research was performed while the first author held a National Research Council Research Associateship Award at the Air Force Research Laboratory. An early version of this paper was presented at the 46th AIAA Aerospace Sciences Meeting and Exhibit, Jan 7–10, 2008, Reno, Nevada.     

Protein crystal growth in microgravity-temperature induced large scale crystallization of insulin

One of the major stumbling blocks that prevents rapid structure determination using x-ray crystallography is macromolecular crystal growth. There are many examples where crystallization takes longer than structure determination. In some cases, it is impossible to grow useful crystals on earth. Recent experiments conducted in conjunction with NASA on various Space Shuttle missions have demonstrated that protein crystals often grow larger and display better internal molecular order than their earth-grown counterparts. This paper reports results from three Shuttle flights using the Protein Crystallization Facility (PCF). The PCF hardware produced large, high-quality insulin crystals by using a temperature change as the sole means to affect protein solubility and thus, crystallization. The facility consists of cylinders/containers with volumes of 500, 200, 100, and 50 ml. Data from the three Shuttle flights demonstrated that larger, higher resolution crystals (as evidenced by x-ray diffraction data) were obtained from the microgravity experiments when compared to earth-grown crystals.     

Viral interleukin 10 (IL-10), the human herpes virus 4 cellular IL-10 homologue, induces local anergy to allogeneic and syngeneic tumors

After the cloning of murine cytokine synthesis inhibitory factor, it was recognized that a homologous open reading frame was encoded within the Epstein-Barr virus (human herpes virus 4). This viral protein has now been termed viral interleukin 10 (vIL-10) to reflect its protein sequence homology to "cellular" IL-10 (cIL-10, either murine or human IL-10). It is now widely accepted that vIL-10 shares many functions with cIL-10, principally, the ability to enhance survival of newly infected B cells and to diminish the production of IFN-gamma and IL-2 during ongoing immune reactions. The immunomodulatory effect of locally secreted vIL-10 and murine IL-10 (mIL-10) was examined in tumor models using CL8-1 (a BL6 melanoma cell line transfected with the H-2Kb class I gene) in syngeneic animals. Although parental BL6 tumor cells grow in immunocompetent syngeneic hosts, CL8-1 are rejected. To achieve local secretion of vIL-10, we generated vIL-10 retroviral vectors. While nontransduced CL8-1 cells (1 x 10(4)) failed to grow when injected intradermally in C57BL/6 mice, CL8-1 cells (1 x 10(4)) transduced with vIL-10 formed palpable tumors and eventually killed 80% of injected animals. Suppression of tumor rejection was also noted when CL8-1 tumors with or without vIL-10 transfection were admixed with syngeneic vIL-10-transfected fibroblasts and inoculated. Since the in vitro proliferation of the tumor was not altered after transduction with the vIL-10 gene and injection of vIL-10-transduced CL8-1 does not affect the rejection of nontransduced CL8-1 inoculated at a distant site, local vIL-10 secretion appears to suppress the process of immune rejection of the target cells in a dose-dependent manner. Similar results were observed for the H-2b MCA105 sarcoma tumor model in allogeneic BALB/c mice (H-2d). Although all animals that received nontransfected MCA105 rapidly rejected these tumors, MCA105 sarcomas transfected with vIL-10 remained palpable for up to 37 d. The local immunosuppressive effect of gene-delivered vIL-10 could be neutralized by anti-human IL-10 monoclonal antibody or could be reversed by the systemic administration of IL-2 or IL-12. In marked contrast, mIL-10 transfection of CL8-1 significantly suppressed tumor growth and frequently led to the rejection of tumor. Similar results were obtained for the murine tumor cell lines MCA102.(ABSTRACT TRUNCATED AT 400 WORDS)     

Occluded Video Instance Segmentation: A Benchmark

International Journal of Computer Vision - Can our video understanding systems perceive objects when a heavy occlusion exists in a scene? To answer this question, we collect a large-scale dataset...     

An approach for robust PDE-constrained optimization with application to shape optimization of electrical engines and of dynamic elastic structures under uncertainty

We present a robust optimization framework that is applicable to general nonlinear programs (NLP) with uncertain parameters. We focus on design problems with partial differential equations (PDE), which involve high computational cost. Our framework addresses the uncertainty with a deterministic worst-case approach. Since the resulting min–max problem is computationally intractable, we propose an approximate robust formulation that employs quadratic models of the involved functions that can be handled efficiently with standard NLP solvers. We outline numerical methods to build the quadratic models, compute their derivatives, and deal with high-dimensional uncertainties. We apply the presented approach to the parametrized shape optimization of systems that are governed by different kinds of PDE and present numerical results.     

The "first generation" of endovascular stent-grafts for patients with aneurysms of the descending thoracic aorta

OBJECTIVE: Our goal was to determine whether endovascular stent-grafting is feasible and effective for patients with aneurysms of the descending thoracic aorta. METHODS: Starting in July 1992, we conducted a prospective, uncontrolled clinical trial in 103 patients (mean age 69 years [range 34-89 years]) who underwent endovascular treatment of aneurysms of the descending thoracic aorta using a custom-fabricated, self-expanding stent-graft device. Follow-up was 100% complete and averaged 22 months. Sixty-two patients (60%) were judged not to be reasonable candidates for a conventional "open" surgical procedure. RESULTS: Complete thrombosis of the aneurysm was ultimately achieved in 86 (83%) patients. The early mortality rate was 9% +/- 3% (+/- 70% CL). Multivariable analysis revealed that myocardial infarction or stroke was linked with a higher likelihood of early death (P = .001). Early serious complications included paraplegia in 3% +/- 2% and stroke in 7% +/- 3%. Actuarial survival estimates at 1 year and 2 years were 81% +/- 4% and 73% +/- 5% (+/- 1 SE), respectively; being judged not to be a surgical candidate portended a higher probability of death (P = .003). According to the intent-to-treat principle, "treatment failure" (including all late sudden unexplained deaths) occurred in 38 patients; 53% +/- 10% of patients were free from treatment failure at 3.7 years. Stent-graft related complications occurred commonly and were linked with several anatomic, technical, and patient-related risk factors. CONCLUSIONS: This 5-year clinical trial involving use of a "first generation" device indicates that endovascular stent-grafting of descending thoracic aortic aneurysms is feasible with acceptable medium-term results. More refined, commercially developed devices available today offer less traumatic and more precise stent-graft deployment; these major technical advantages, coupled with important lessons we have learned over time and better patient selection, should be associated with more salutary clinical results in the future.