Trace-based Analysis Research Articles

Cross-User Similarities in Viewing Behavior for 360° Video and Caching Implications

The demand and usage of 360° video services are expected to increase. However, despite these services being highly bandwidth intensive, not much is known about the potential value that basic bandwidth saving techniques such as server or edge-network on-demand caching (e.g., in a CDN) could have when used for delivery of such services. This problem is both important and complicated as client-side solutions have been developed that split the full 360° view into multiple tiles, and adapt the quality of the downloaded tiles based on the user’s expected viewing direction and bandwidth conditions. This article presents new trace-based analysis methods that incorporate users’ viewports (the area of the full 360° view the user actually sees), a first characterization of the cross-user similarities of the users’ viewports, and a trace-based analysis of the potential bandwidth savings that caching-based techniques may offer under different conditions. Our analysis takes into account differences in the time granularity over which viewport overlaps can be beneficial for resource saving techniques, compares and contrasts differences between video categories, and accounts for uncertainties in the network conditions and the prediction of the future viewing direction when prefetching. The results provide substantial insight into the conditions under which overlap can be considerable and caching effective, and inform the design of new caching system policies tailored for 360° video.

Geolocation-based sector selection for Vehicle-to-Infrastructure 802.11ad communication

To improve range, IEEE 802.11ad uses directional communication, the first step of which is to choose the best antenna configuration, known as sector. We studied the sector selection behavior of Commercial Off-The-Shelf (COTS) 802.11ad equipment in an experimental Vehicle-to-Infrastructure (V2I) communication scenario. First, we created a framework for mm-Wave data collection in mobile scenarios, highlighting the challenges, justifying our solutions, and making them available, thus flattening the path for future experimentation by others. We then proceeded to use said framework to collect data in a realistic V2I communication scenario. We performed two independent measurement campaigns in the same area. Analysis of the collected data revealed the following inefficiencies in the devices’ sector selection algorithm: (i) a large number of sector selection attempts that do not result in a sector change; and (ii) a “ping-pong” effect in which a node oscillates between two sectors. With this in mind we propose an alternative antenna sector selection scheme that uses spatially-indexed historical performance data to pick the statistically-best sector for any given geolocation. A trace-based analysis showed that a position-based strategy can improve throughput by more than 10% for 30% of locations, and that gains can be as high as 60%, in some instances.

Locality analysis through static parallel sampling

Locality analysis is important since accessing memory is much slower than computing. Compile-time locality analysis can provide detailed program-level feedback for compilers or runtime systems faster than trace-based locality analysis. In this paper, we describe a new approach to locality analysis based on static parallel sampling. A compiler analyzes loop-based code and generates sampler code which is run to measure locality. Our approach can predict precise cache line granularity miss ratio curves for complex loops with non-linear array references and even branches. The precision and overhead of static sampling are evaluated using PolyBench and a bit-reversal loop. Our result shows that by randomly sampling 2% of loop iterations, a compiler can construct almost exact miss ratio curves as trace based analysis. Sampling 0.5% and 1% iterations can achieve good precision and efficiency with an average 0.6% to 1% the time of tracing respectively. Our analysis can also be parallelized. The analysis may assist program optimization techniques such as tiling, program co-location, cache hint selection and help to analyze write locality and parallel locality.

Fast classification of MPI applications using Lamport’s logical clocks

We present a novel trace-based analysis tool that rapidly classifies an MPI application as bandwidth-bound, latency-bound, load-imbalance-bound, or computation-bound for different interconnection networks. The tool uses an extension of Lamport’s logical clock to track application progress in the trace replay. It has two unique features. First, it can predict application performance for many latency and bandwidth parameters from a single replay of the trace. Second, it infers the performance characteristics of an application and classifies the application using the predicted performance trend for a range of network configurations instead of using the predicted performance for a particular network configuration. We describe the techniques used in the tool and its design and implementation, and report our performance study of the tool and our experience with classifying twelve applications and mini-apps from the DOE DesignForward project as well as the NAS Parallel Benchmarks.

Fog Caching and a Trace-Based Analysis of its Offload Effect

Many years of research on Content Delivery Networks (CDNs) offers a number of effective methods for caching of content replicas or forwarding requests. However, recently CDNs have aggressively started migrating to clouds. Clouds present a new kind of distribution environment as each location can support multiple caching options varying in the level of persistence of stored content. A subclass of clouds located at network edge is referred to as fog clouds. Fog clouds help by allowing CDNs to offload popular content to network edge, closer to end users. However, due to the fact that fog clouds are extremely heterogeneous and vary wildly in network and caching performance, traditional caching technology is no longer applicable. This paper proposes a multi-level caching technology specific to fog clouds. To deal with the heterogeneity problem and, at the same time, avoid centralized control, this paper proposes a function that allows CDN services to discover local caching facilities dynamically, at runtime. Using a combination of synthetic models and real measurement dataset, this paper analyzes efficiency of offload both at the local level of individual fog locations and at the global level of the entire CDN infrastructure. Local analysis shows that the new method can reduce inter-cloud traffic by between 16 and 18 times while retaining less than 30% of total content in a local cache. Global analysis further shows that, based on existing measurement datasets, centralized optimization is preferred to distributed coordination among services.

Forest logging: A trace-based analysis of large rule-based computations

Knowledge representation systems based on the well-founded semantics can offer the degree of scalability required for semantic web applications and make use of ex- pressive semantic features such as Hilog, frame-based rea- soning, and defeasibility theories. Such features can be com- piled into Prolog tabling engines that have good support for indexing and memory management. However, due both to the power of the semantic features and to the declarative style typical of knowledge representation rules, the resources needed for query evaluation can be unpredictable. In such a situation, users need to understand the overall structure of a computation and examine problematic portions of it. This problem, of profiling a computation, differs from debug- ging and justification which address why a given answer was or wasn't derived, and so profiling requires different tech- niques. In this paper we present a trace-based analysis tech- nique called forest logging which has been used to profile large, heavily tabled computations. In forest logging, criti- cal aspects of a tabled computation are logged; afterwards the log is loaded and analyzed. As implemented in XSB, for- est logging slows down execution of practical programs by a constant factor that is often small; and logs containing tens or hundreds of millions of facts can be loaded and analyzed in minutes.

Resource Provisioning for Routing with Priorities

In this paper, a novel load balancing technique is proposed to handle biased call request patterns efficiently. The knowledge of call request patterns is used from trace-based analysis; call patterns are affected by various factors such as geographical context, user mobility, network usage patterns and temporal bias. A call routing system is modelled as a network of queues and a provisioning algorithm is developed. The proposed model employs a combination of predictive and reactive provisioning methods. The idea of Capacity Distribution is introduced- a heuristic for heterogeneous capacity allocation among adjacent cell pairs. The proposed model retains simplicity while being able to effectively learn from the variations in call patterns. The performance of the model is evaluated using extensive simulation techniques.

Dynamic trace-based analysis of vectorization potential of applications

Recent hardware trends with GPUs and the increasing vector lengths of SSE-like ISA extensions for multicore CPUs imply that effective exploitation of SIMD parallelism is critical for achieving high performance on emerging and future architectures. A vast majority of existing applications were developed without any attention by their developers towards effective vectorizability of the codes. While developers of production compilers such as GNU gcc, Intel icc, PGI pgcc, and IBM xlc have invested considerable effort and made significant advances in enhancing automatic vectorization capabilities, these compilers still cannot effectively vectorize many existing scientific and engineering codes. It is therefore of considerable interest to analyze existing applications to assess the inherent latent potential for SIMD parallelism, exploitable through further compiler advances and/or via manual code changes. In this paper we develop an approach to infer a program's SIMD parallelization potential by analyzing the dynamic data-dependence graph derived from a sequential execution trace. By considering only the observed run-time data dependences for the trace, and by relaxing the execution order of operations to allow any dependence-preserving reordering, we can detect potential SIMD parallelism that may otherwise be missed by more conservative compile-time analyses. We show that for several benchmarks our tool discovers regions of code within computationally-intensive loops that exhibit high potential for SIMD parallelism but are not vectorized by state-of-the-art compilers. We present several case studies of the use of the tool, both in identifying opportunities to enhance the transformation capabilities of vectorizing compilers, as well as in pointing to code regions to manually modify in order to enable auto-vectorization and performance improvement by existing compilers.

Analysis of information flow security in cyber–physical systems

Abstract Information flow is a fundamental concept underlying the security of a system. Confidentiality of information in a system can be breached through unrestricted information flow. Physical components added to a cyber system considerably increase the difficulty of determining information flow and the difficulty of mitigating the corresponding confidentiality problem. Fundamentally, physical actions inherently divulge information through simple observation. This work applies classical models of non-deducibility and non-inference to cyber–physical systems (CPSs) to determine information flow in the coupled cyber and physical worlds. The results demonstrate that the combined physical and cyber properties of a CPS can both protect and divulge information. The key to formalizing this analysis is to find a uniform semantic representation of the cyber and physical components, their interaction, and the physics of the system, and to devise a formal modeling technique for determining information flow. This paper presents a semantic model for information flow analysis in a CPS and describes an approach to perform the analysis, including both trace-based analysis and automated analysis through process algebra specification. Two model infrastructures demonstrate the approach, a gas pipeline system and a smart electric power grid system. In the gas pipeline system, cooperating flow control system devices exchange confidential information to produce physical actions in the pipeline. In the smart grid, cooperating flexible alternating current transmission system devices exchange confidential information to produce physical actions. The approach can verify whether these infrastructures inherently preserve confidentiality.

Trace-based analysis of Wi-Fi scanning strategies

The number of smartphones in use is overwhelmingly increasing every year. These devices rely on connectivity to the Internet for the majority of their applications. The everincreasing number of deployed 802.11 wireless access points and the relatively high cost of other data services make the case for opportunistic communication using free WiFi hotspots. However, this requires effective management of the WLAN interface, because by design the energy cost of WLAN scanning and interface idle operation is high and energy is a primary resource on mobile devices. We study several heuristic strategies for interface management, and use real-world user traces to evaluate and compare their performance against the optimal algorithm. Trace-based simulations show that simple static scanning with a suitable interval value is very effective for delay-tolerant, background applications.

Analysis of Load Balancing Issues Caused by Intra-Movie Skewness for Parallel Video Servers

Parallel video servers divide movies into segments and put them across multiple nodes, which avoid load imbalance problem caused by movie popularity. However, due to users' various viewing time, the retrieval rates of movie segments are quite different, and some segments are more popular than others. This instance is called intra-movie skewness, which probably leads to load imbalance among the nodes in parallel video servers. In this paper we first study intra-movie skewness through a trace-based analysis. It is proved that this skewness results in serious performance degradation of the video server with traditional round robin data placement. Then, according to the characteristics of intra-movie skewness, we propose a strategy, Symmetrical Pair Scheme to introduce our symmetrical data placement style, which has better load-balancing performance. Moreover, based on Symmetrical Pair Scheme we present a novel method, Symmetrical Replication Pair Scheme. Our analysis result shows that it can prevent the impact of intra-movie skewness and bring desirable load-balancing performance to parallel video servers.

On the scale and performance of cooperative Web proxy caching

While algorithms for cooperative proxy caching have been widely studied, little is understood about cooperative-caching performance in the large-scale World Wide Web environment. This paper uses both trace-based analysis and analytic modelling to show the potential advantages and drawbacks of inter-proxy cooperation. With our traces, we evaluate quantitatively the performance-improvement potential of cooperation between 200 small-organization proxies within a university environment, and between two large-organization proxies handling 23,000 and 60,000 clients, respectively. With our model, we extend beyond these populations to project cooperative caching behavior in regions with millions of clients. Overall, we demonstrate that cooperative caching has performance benefits only within limited population bounds. We also use our model to examine the implications of future trends in Web-access behavior and traffic.