skip to main content


Title: PerfDebug: Performance Debugging of Computation Skew in Dataflow Systems
Performance is a key factor for big data applications, and much research has been devoted to optimizing these applications. While prior work can diagnose and correct data skew, the problem of computation skew---abnormally high computation costs for a small subset of input data---has been largely overlooked. Computation skew commonly occurs in real-world applications and yet no tool is available for developers to pinpoint underlying causes. To enable a user to debug applications that exhibit computation skew, we develop a post-mortem performance debugging tool. PerfDebug automatically finds input records responsible for such abnormalities in a big data application by reasoning about deviations in performance metrics such as job execution time, garbage collection time, and serialization time. The key to PerfDebug's success is a data provenance-based technique that computes and propagates record-level computation latency to keep track of abnormally expensive records throughout the pipeline. Finally, the input records that have the largest latency contributions are presented to the user for bug fixing. We evaluate PerfDebug via in-depth case studies and observe that remediation such as removing the single most expensive record or simple code rewrite can achieve up to 16X performance improvement.  more » « less
Award ID(s):
1764077 1763172
NSF-PAR ID:
10173701
Author(s) / Creator(s):
; ; ;
Date Published:
Journal Name:
SoCC '19: Proceedings of the ACM Symposium on Cloud Computing
Page Range / eLocation ID:
465 to 476
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Obeid, Iyad Selesnick (Ed.)
    Electroencephalography (EEG) is a popular clinical monitoring tool used for diagnosing brain-related disorders such as epilepsy [1]. As monitoring EEGs in a critical-care setting is an expensive and tedious task, there is a great interest in developing real-time EEG monitoring tools to improve patient care quality and efficiency [2]. However, clinicians require automatic seizure detection tools that provide decisions with at least 75% sensitivity and less than 1 false alarm (FA) per 24 hours [3]. Some commercial tools recently claim to reach such performance levels, including the Olympic Brainz Monitor [4] and Persyst 14 [5]. In this abstract, we describe our efforts to transform a high-performance offline seizure detection system [3] into a low latency real-time or online seizure detection system. An overview of the system is shown in Figure 1. The main difference between an online versus offline system is that an online system should always be causal and has minimum latency which is often defined by domain experts. The offline system, shown in Figure 2, uses two phases of deep learning models with postprocessing [3]. The channel-based long short term memory (LSTM) model (Phase 1 or P1) processes linear frequency cepstral coefficients (LFCC) [6] features from each EEG channel separately. We use the hypotheses generated by the P1 model and create additional features that carry information about the detected events and their confidence. The P2 model uses these additional features and the LFCC features to learn the temporal and spatial aspects of the EEG signals using a hybrid convolutional neural network (CNN) and LSTM model. Finally, Phase 3 aggregates the results from both P1 and P2 before applying a final postprocessing step. The online system implements Phase 1 by taking advantage of the Linux piping mechanism, multithreading techniques, and multi-core processors. To convert Phase 1 into an online system, we divide the system into five major modules: signal preprocessor, feature extractor, event decoder, postprocessor, and visualizer. The system reads 0.1-second frames from each EEG channel and sends them to the feature extractor and the visualizer. The feature extractor generates LFCC features in real time from the streaming EEG signal. Next, the system computes seizure and background probabilities using a channel-based LSTM model and applies a postprocessor to aggregate the detected events across channels. The system then displays the EEG signal and the decisions simultaneously using a visualization module. The online system uses C++, Python, TensorFlow, and PyQtGraph in its implementation. The online system accepts streamed EEG data sampled at 250 Hz as input. The system begins processing the EEG signal by applying a TCP montage [8]. Depending on the type of the montage, the EEG signal can have either 22 or 20 channels. To enable the online operation, we send 0.1-second (25 samples) length frames from each channel of the streamed EEG signal to the feature extractor and the visualizer. Feature extraction is performed sequentially on each channel. The signal preprocessor writes the sample frames into two streams to facilitate these modules. In the first stream, the feature extractor receives the signals using stdin. In parallel, as a second stream, the visualizer shares a user-defined file with the signal preprocessor. This user-defined file holds raw signal information as a buffer for the visualizer. The signal preprocessor writes into the file while the visualizer reads from it. Reading and writing into the same file poses a challenge. The visualizer can start reading while the signal preprocessor is writing into it. To resolve this issue, we utilize a file locking mechanism in the signal preprocessor and visualizer. Each of the processes temporarily locks the file, performs its operation, releases the lock, and tries to obtain the lock after a waiting period. The file locking mechanism ensures that only one process can access the file by prohibiting other processes from reading or writing while one process is modifying the file [9]. The feature extractor uses circular buffers to save 0.3 seconds or 75 samples from each channel for extracting 0.2-second or 50-sample long center-aligned windows. The module generates 8 absolute LFCC features where the zeroth cepstral coefficient is replaced by a temporal domain energy term. For extracting the rest of the features, three pipelines are used. The differential energy feature is calculated in a 0.9-second absolute feature window with a frame size of 0.1 seconds. The difference between the maximum and minimum temporal energy terms is calculated in this range. Then, the first derivative or the delta features are calculated using another 0.9-second window. Finally, the second derivative or delta-delta features are calculated using a 0.3-second window [6]. The differential energy for the delta-delta features is not included. In total, we extract 26 features from the raw sample windows which add 1.1 seconds of delay to the system. We used the Temple University Hospital Seizure Database (TUSZ) v1.2.1 for developing the online system [10]. The statistics for this dataset are shown in Table 1. A channel-based LSTM model was trained using the features derived from the train set using the online feature extractor module. A window-based normalization technique was applied to those features. In the offline model, we scale features by normalizing using the maximum absolute value of a channel [11] before applying a sliding window approach. Since the online system has access to a limited amount of data, we normalize based on the observed window. The model uses the feature vectors with a frame size of 1 second and a window size of 7 seconds. We evaluated the model using the offline P1 postprocessor to determine the efficacy of the delayed features and the window-based normalization technique. As shown by the results of experiments 1 and 4 in Table 2, these changes give us a comparable performance to the offline model. The online event decoder module utilizes this trained model for computing probabilities for the seizure and background classes. These posteriors are then postprocessed to remove spurious detections. The online postprocessor receives and saves 8 seconds of class posteriors in a buffer for further processing. It applies multiple heuristic filters (e.g., probability threshold) to make an overall decision by combining events across the channels. These filters evaluate the average confidence, the duration of a seizure, and the channels where the seizures were observed. The postprocessor delivers the label and confidence to the visualizer. The visualizer starts to display the signal as soon as it gets access to the signal file, as shown in Figure 1 using the “Signal File” and “Visualizer” blocks. Once the visualizer receives the label and confidence for the latest epoch from the postprocessor, it overlays the decision and color codes that epoch. The visualizer uses red for seizure with the label SEIZ and green for the background class with the label BCKG. Once the streaming finishes, the system saves three files: a signal file in which the sample frames are saved in the order they were streamed, a time segmented event (TSE) file with the overall decisions and confidences, and a hypotheses (HYP) file that saves the label and confidence for each epoch. The user can plot the signal and decisions using the signal and HYP files with only the visualizer by enabling appropriate options. For comparing the performance of different stages of development, we used the test set of TUSZ v1.2.1 database. It contains 1015 EEG records of varying duration. The any-overlap performance [12] of the overall system shown in Figure 2 is 40.29% sensitivity with 5.77 FAs per 24 hours. For comparison, the previous state-of-the-art model developed on this database performed at 30.71% sensitivity with 6.77 FAs per 24 hours [3]. The individual performances of the deep learning phases are as follows: Phase 1’s (P1) performance is 39.46% sensitivity and 11.62 FAs per 24 hours, and Phase 2 detects seizures with 41.16% sensitivity and 11.69 FAs per 24 hours. We trained an LSTM model with the delayed features and the window-based normalization technique for developing the online system. Using the offline decoder and postprocessor, the model performed at 36.23% sensitivity with 9.52 FAs per 24 hours. The trained model was then evaluated with the online modules. The current performance of the overall online system is 45.80% sensitivity with 28.14 FAs per 24 hours. Table 2 summarizes the performances of these systems. The performance of the online system deviates from the offline P1 model because the online postprocessor fails to combine the events as the seizure probability fluctuates during an event. The modules in the online system add a total of 11.1 seconds of delay for processing each second of the data, as shown in Figure 3. In practice, we also count the time for loading the model and starting the visualizer block. When we consider these facts, the system consumes 15 seconds to display the first hypothesis. The system detects seizure onsets with an average latency of 15 seconds. Implementing an automatic seizure detection model in real time is not trivial. We used a variety of techniques such as the file locking mechanism, multithreading, circular buffers, real-time event decoding, and signal-decision plotting to realize the system. A video demonstrating the system is available at: https://www.isip.piconepress.com/projects/nsf_pfi_tt/resources/videos/realtime_eeg_analysis/v2.5.1/video_2.5.1.mp4. The final conference submission will include a more detailed analysis of the online performance of each module. ACKNOWLEDGMENTS Research reported in this publication was most recently supported by the National Science Foundation Partnership for Innovation award number IIP-1827565 and the Pennsylvania Commonwealth Universal Research Enhancement Program (PA CURE). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the official views of any of these organizations. REFERENCES [1] A. Craik, Y. He, and J. L. Contreras-Vidal, “Deep learning for electroencephalogram (EEG) classification tasks: a review,” J. Neural Eng., vol. 16, no. 3, p. 031001, 2019. https://doi.org/10.1088/1741-2552/ab0ab5. [2] A. C. Bridi, T. Q. Louro, and R. C. L. Da Silva, “Clinical Alarms in intensive care: implications of alarm fatigue for the safety of patients,” Rev. Lat. Am. Enfermagem, vol. 22, no. 6, p. 1034, 2014. https://doi.org/10.1590/0104-1169.3488.2513. [3] M. Golmohammadi, V. Shah, I. Obeid, and J. Picone, “Deep Learning Approaches for Automatic Seizure Detection from Scalp Electroencephalograms,” in Signal Processing in Medicine and Biology: Emerging Trends in Research and Applications, 1st ed., I. Obeid, I. Selesnick, and J. Picone, Eds. New York, New York, USA: Springer, 2020, pp. 233–274. https://doi.org/10.1007/978-3-030-36844-9_8. [4] “CFM Olympic Brainz Monitor.” [Online]. Available: https://newborncare.natus.com/products-services/newborn-care-products/newborn-brain-injury/cfm-olympic-brainz-monitor. [Accessed: 17-Jul-2020]. [5] M. L. Scheuer, S. B. Wilson, A. Antony, G. Ghearing, A. Urban, and A. I. Bagic, “Seizure Detection: Interreader Agreement and Detection Algorithm Assessments Using a Large Dataset,” J. Clin. Neurophysiol., 2020. https://doi.org/10.1097/WNP.0000000000000709. [6] A. Harati, M. Golmohammadi, S. Lopez, I. Obeid, and J. Picone, “Improved EEG Event Classification Using Differential Energy,” in Proceedings of the IEEE Signal Processing in Medicine and Biology Symposium, 2015, pp. 1–4. https://doi.org/10.1109/SPMB.2015.7405421. [7] V. Shah, C. Campbell, I. Obeid, and J. Picone, “Improved Spatio-Temporal Modeling in Automated Seizure Detection using Channel-Dependent Posteriors,” Neurocomputing, 2021. [8] W. Tatum, A. Husain, S. Benbadis, and P. Kaplan, Handbook of EEG Interpretation. New York City, New York, USA: Demos Medical Publishing, 2007. [9] D. P. Bovet and C. Marco, Understanding the Linux Kernel, 3rd ed. O’Reilly Media, Inc., 2005. https://www.oreilly.com/library/view/understanding-the-linux/0596005652/. [10] V. Shah et al., “The Temple University Hospital Seizure Detection Corpus,” Front. Neuroinform., vol. 12, pp. 1–6, 2018. https://doi.org/10.3389/fninf.2018.00083. [11] F. Pedregosa et al., “Scikit-learn: Machine Learning in Python,” J. Mach. Learn. Res., vol. 12, pp. 2825–2830, 2011. https://dl.acm.org/doi/10.5555/1953048.2078195. [12] J. Gotman, D. Flanagan, J. Zhang, and B. Rosenblatt, “Automatic seizure detection in the newborn: Methods and initial evaluation,” Electroencephalogr. Clin. Neurophysiol., vol. 103, no. 3, pp. 356–362, 1997. https://doi.org/10.1016/S0013-4694(97)00003-9. 
    more » « less
  2. null (Ed.)
    Debugging big data analytics often requires a root cause analysis to pinpoint the precise culprit records in an input dataset responsible for incorrect or anomalous output. Existing debugging or data provenance approaches do not track fine-grained control and data flows in user-defined application code; thus, the returned culprit data is often too large for manual inspection and expensive post-mortem analysis is required. We design FlowDebug to identify a highly precise set of input records based on two key insights. First, FlowDebug precisely tracks control and data flow within user-defined functions to propagate taints at a fine-grained level by inserting custom data abstractions through automated source to source transformation. Second, it introduces a novel notion of influence-based provenance for many-to-one dependencies to prioritize which input records are more responsible than others by analyzing the semantics of a user-defined function used for aggregation. By design, our approach does not require any modification to the framework's runtime and can be applied to existing applications easily. FlowDebug significantly improves the precision of debugging results by up to 99.9 percentage points and avoids repetitive re-runs required for post-mortem analysis by a factor of 33 while incurring an instrumentation overhead of 0.4X - 6.1X on vanilla Spark. 
    more » « less
  3. Binder is a publicly accessible online service for executing interactive notebooks based on Git repositories. Binder dynamically builds and deploys containers following a recipe stored in the repository, then gives the user a browser-based notebook interface. The Binder group periodically releases a log of container launches from the public Binder service. Archives of launch records are available here. These records do not include identifiable information like IP addresses, but do give the source repo being launched along with some other metadata. The main content of this dataset is in the binder.sqlite file. This SQLite database includes launch records from 2018-11-03 to 2021-06-06 in the events table, which has the following schema.

    CREATE TABLE events( version INTEGER, timestamp TEXT, provider TEXT, spec TEXT, origin TEXT, ref TEXT, guessed_ref TEXT ); CREATE INDEX idx_timestamp ON events(timestamp);
    • version indicates the version of the record as assigned by Binder. The origin field became available with version 3, and the ref field with version 4. Older records where this information was not recorded will have the corresponding fields set to null.
    • timestamp is the ISO timestamp of the launch
    • provider gives the type of source repo being launched ("GitHub" is by far the most common). The rest of the explanations assume GitHub, other providers may differ.
    • spec gives the particular branch/release/commit being built. It consists of <github-id>/<repo>/<branch>.
    • origin indicates which backend was used. Each has its own storage, compute, etc. so this info might be important for evaluating caching and performance. Note that only recent records include this field. May be null.
    • ref specifies the git commit that was actually used, rather than the named branch referenced by spec. Note that this was not recorded from the beginning, so only the more recent entries include it. May be null.
    • For records where ref is not available, we attempted to clone the named reference given by spec rather than the specific commit (see below). The guessed_ref field records the commit found at the time of cloning. If the branch was updated since the container was launched, this will not be the exact version that was used, and instead will refer to whatever was available at the time (early 2021). Depending on the application, this might still be useful information. Selecting only records with version 4 (or non-null ref) will exclude these guessed commits. May be null.

    The Binder launch dataset identifies the source repos that were used, but doesn't give any indication of their contents. We crawled GitHub to get the actual specification files in the repos which were fed into repo2docker when preparing the notebook environments, as well as filesystem metadata of the repos. Some repos were deleted/made private at some point, and were thus skipped. This is indicated by the absence of any row for the given commit (or absence of both ref and guessed_ref in the events table). The schema is as follows.

    CREATE TABLE spec_files ( ref TEXT NOT NULL PRIMARY KEY, ls TEXT, runtime BLOB, apt BLOB, conda BLOB, pip BLOB, pipfile BLOB, julia BLOB, r BLOB, nix BLOB, docker BLOB, setup BLOB, postbuild BLOB, start BLOB );

    Here ref corresponds to ref and/or guessed_ref from the events table. For each repo, we collected spec files into the following fields (see the repo2docker docs for details on what these are). The records in the database are simply the verbatim file contents, with no parsing or further processing performed.

    • runtime: runtime.txt
    • apt: apt.txt
    • conda: environment.yml
    • pip: requirements.txt
    • pipfile: Pipfile.lock or Pipfile
    • julia: Project.toml or REQUIRE
    • r: install.R
    • nix: default.nix
    • docker: Dockerfile
    • setup: setup.py
    • postbuild: postBuild
    • start: start

    The ls field gives a metadata listing of the repo contents (excluding the .git directory). This field is JSON encoded with the following structure based on JSON types:

    • Object: filesystem directory. Keys are file names within it. Values are the contents, which can be regular files, symlinks, or subdirectories.
    • String: symlink. The string value gives the link target.
    • Number: regular file. The number value gives the file size in bytes.
    CREATE TABLE clean_specs ( ref TEXT NOT NULL PRIMARY KEY, conda_channels TEXT, conda_packages TEXT, pip_packages TEXT, apt_packages TEXT );

    The clean_specs table provides parsed and validated specifications for some of the specification files (currently Pip, Conda, and APT packages). Each column gives either a JSON encoded list of package requirements, or null. APT packages have been validated using a regex adapted from the repo2docker source. Pip packages have been parsed and normalized using the Requirement class from the pkg_resources package of setuptools. Conda packages have been parsed and normalized using the conda.models.match_spec.MatchSpec class included with the library form of Conda (distinct from the command line tool). Users might want to use these parsers when working with the package data, as the specifications can become fairly complex.

    The missing table gives the repos that were not accessible, and event_logs records which log files have already been added. These tables are used for updating the dataset and should not be of interest to users.

     
    more » « less
  4. We present TVA, a multi-party computation (MPC) system for secure analytics on secret-shared time series data. TVA achieves strong security guarantees in the semi-honest and malicious settings, and high expressivity by enabling complex analytics on inputs with unordered and irregular timestamps. TVA is the first system to support arbitrary composition of oblivious window operators, keyed aggregations, and multiple filter predicates, while keeping all data attributes private, including record timestamps and user-defined values in query predicates. At the core of the TVA system lie novel protocols for secure window assignment: (i) a tumbling window protocol that groups records into fixed-length time buckets and (ii) two session window protocols that identify periods of activity followed by periods of inactivity. We also contribute a new protocol for secure division with a public divisor, which may be of independent interest. We evaluate TVA on real LAN and WAN environments and show that it can efficiently compute complex window-based analytics on inputs of 2^22 records with modest use of resources. When compared to the state-of-the-art, TVA achieves up to 5.8× lower latency in queries with multiple filters and two orders of magnitude better performance in window aggregation. 
    more » « less
  5. null (Ed.)
    As big data analytics become increasingly popular, data-intensive scalable computing (DISC) systems help address the scalability issue of handling large data. However, automated testing for such data-centric applications is challenging, because data is often incomplete, continuously evolving, and hard to know a priori. Fuzz testing has been proven to be highly effective in other domains such as security; however, it is nontrivial to apply such traditional fuzzing to big data analytics directly for three reasons: (1) the long latency of DISC systems prohibits the applicability of fuzzing: naïve fuzzing would spend 98% of the time in setting up a test environment; (2) conventional branch coverage is unlikely to scale to DISC applications because most binary code comes from the framework implementation such as Apache Spark; and (3) random bit or byte level mutations can hardly generate meaningful data, which fails to reveal real-world application bugs. We propose a novel coverage-guided fuzz testing tool for big data analytics, called BigFuzz. The key essence of our approach is that: (a) we focus on exercising application logic as opposed to increasing framework code coverage by abstracting the DISC framework using specifications. BigFuzz performs automated source to source transformations to construct an equivalent DISC application suitable for fast test generation, and (b) we design schema-aware data mutation operators based on our in-depth study of DISC application error types. BigFuzz speeds up the fuzzing time by 78 to 1477X compared to random fuzzing, improves application code coverage by 20% to 271%, and achieves 33% to 157% improvement in detecting application errors. When compared to the state of the art that uses symbolic execution to test big data analytics, BigFuzz is applicable to twice more programs and can find 81% more bugs. 
    more » « less