End-to-End Big Data AI with Analytics Zoo

End-to-End Big Data AI with Analytics Zoo
End-to-End Big Data AI with Analytics Zoo Jason Dai – Sr. Principal Engineer Fadi Zuhayri – Sr. Director Intel Architecture, Graphics & Software
• Intel’s Transformation for Intelligence Era • Analytics Zoo: Software Platform for Big Data AI • Building Big Data AI Applications on Analytics Zoo Outline
Fadi Zuhayri Sr. Director
COMPUTE 1980 1990 2000 2010 2020 2030 2040 PC ERA DIGITIZE EVERYTHING NETWORK EVERYTHING 1 BILLION INTERNET CONNECTED DEVICES 1018 109 104 1015 102 T E C H N O L O G Y L E D D I S R U P T I O N S COMPUTE DEMOCRATIZATION
COMPUTE 1980 1990 2000 2010 2020 2030 2040 PC ERA DIGITIZE EVERYTHING NETWORK EVERYTHING 1 BILLION INTERNET CONNECTED DEVICES 1018 109 104 1015 102 CLOUD EVERYTHING MOBILE EVERYTHING MOBILE + CLOUD ERA 10 BILLION CLOUD CONNECTED DEVICES T E C H N O L O G Y L E D D I S R U P T I O N S COMPUTE DEMOCRATIZATION
COMPUTE 1980 1990 2000 2010 2020 2030 2040 PC ERA DIGITIZE EVERYTHING NETWORK EVERYTHING 1 BILLION INTERNET CONNECTED DEVICES 1018 109 104 1015 102 100 BILLION INTELLIGENT CONNECTED DEVICES INTELLIGENCE ERA CLOUD EVERYTHING MOBILE EVERYTHING MOBILE + CLOUD ERA 10 BILLION CLOUD CONNECTED DEVICES T E C H N O L O G Y L E D D I S R U P T I O N S COMPUTE DEMOCRATIZATION
COMPUTE 1980 1990 2000 2010 2020 2030 2040 PC ERA DIGITIZE EVERYTHING NETWORK EVERYTHING 1 BILLION INTERNET CONNECTED DEVICES 1018 109 104 1015 102 T E C H N O L O G Y L E D D I S R U P T I O N S COMPUTE DEMOCRATIZATION F O R E V E R Y O N E EXASCALE CLOUD EVERYTHING MOBILE EVERYTHING MOBILE + CLOUD ERA 10 BILLION CLOUD CONNECTED DEVICES
Industry inflections are fueling the growth of data Intelligent Edge Artificial Intelligence 5G Network Transformation Cloudification
Move Faster Store More Process Everything Software & System Level Optimized Intel® Silicon Photonics Intel® Ethernet Intel® Tofino Unleashing the Potential of Data
Analytics & AI Strategy 21 4 3 Hardware Software Ecosystem OPTIMIZED SOFTWARE E2E DATA SCIENCE UNIFIED APIs CPU INFUSED WITH AI FLEXIBLE ACCELERATION OPTIMIZED PLATFORM A THRIVING COMMUNITY INTELLIGENT SOLUTIONS INNOVATION & INVESTMENT
XPU: DIVERSE INTEL HW PORTFOLIO – from edge to cloud CPU General compute, AI Inference & training Xe GPU HPC & AI AI training & inference Habana Low power vision Low power NLPGNA Movidius
24 OPTIMIZED TOPOLOGIES 44 OPTIMIZED TOPOLOGIES 100+ OPTIMIZED TOPOLOGIES … Foundation for AI 13 More built-in AI acceleration & optimized topologies with each new gen OPTIMIZED LIBRARIES AND FRAMEWORKS 2017 1ST GEN Intel® Advanced Vector Extensions 512 (Intel AVX-512) 2019 2ND GEN Intel Deep Learning Boost (with VNNI) 2020 3RD GEN Intel Deep Learning Boost (VNNI, BF16) Intel Deep Learning Boost (AMX) 2021 NEXT GEN AIPERFORMANCE CPU INFUSED WITH AI
Languages and Libraries Middleware & Frameworks Application Workloads CPU GPU FPGAOther Accelerators Scalar Vector Matrix Spatial
XPUs Languages and Libraries Middleware & Frameworks Application Workloads CPU GPU FPGAOther Accelerators Scalar Vector Matrix Spatial
2020 2021 Q1 Q2 Q3 Q4Q4Q3 0.6 Spec. 0.7 Spec. 0.8 Spec. 0.9 Spec. 1.0 Spec. Industry Initiative Announced More Soon… Learn more at oneapi.com
Middleware, Frameworks & Runtimes Applications & Services XPUs CPU GPU FPGA Intel® oneAPI Product Gold Available December 2020 ... Compatibility Tool Languages Libraries Analysis & Debug Tools Hardware Abstraction Layer
Run Locally Run in the Cloud Get started quickly: code samples, quick-start guides, webinars, training software.intel.com/oneapi Downloads Repositories Containers DevCloud
One Minute to Code No Hardware Acquisition No Download, Install or Configuration Support for Jupyter Notebooks, VS Code Easy Access to Samples and Tutorials
ECOSYSTEM OF AI SOFTWARE STACK HW LIBRARIES & COMPILERS AI/ANALYTICS SOLUTIONS DL/ML/BIGDATA FRAMEWORKS XPU oneDNN CPU oneDAL oneCCL DATA SCIENTISTS & DATA ANALYSTS GPU ACCELERATERS M O D E L Z O O A N A L Y T I C S Z O O O P E N - V I N O ™ T E N S O R - F L O W P Y T H O N / N U M B A T V M P Y - T O R C H M X N E T S P A R K S Q L + M L / DL S c a l e O ut M O D I N N U M P Y X G - BO O S T S C I K I T - L E A R N P A N D A S
21 Achieving higher yields and efficiency Increasing production and uptime Transforming learning Enhancing safety Revolutionizing patient outcomes Turning data into value Pervasive Analytics & AI Agriculture Energy Education Government Finance Healthcare Empowering industry 4.0 Creating thrilling experiences Modernizing shopping Enabling homes to see, hear & respond Fueling automated driving Driving network efficiency Industrial Media Retail Smart Home Telecom Transport intel.com/customerspotlight
INFERENCETRAINING FEATURE ENGINEERING AI TREND: END TO END GROWING DEMAND FOR END-TO-END AI PIPELINE DATA INGESTION
Big Data AI Open-Source Software Platform Distributed TensorFlow, PyTorch, Keras, BigDL, RAY, and Apache Spark Reference Use Cases, AI Models, High-level APIs, Feature Engineering, etc. https://github.com/intel-analytics/analytics-zoo AI on Big Data Simplifying End-to-End Big Data AI Solutions Development
Jason Dai Senior Principal Engineer
• Big Data AI • Analytics Zoo overview • Summary Agenda
Seamless Scaling from Laptop to Distributed Big Data Distributed, High-Performance Deep Learning Framework for Apache Spark https://github.com/intel-analytics/bigdl Unified Big Data AI Platform for TensorFlow, PyTorch, Keras, BigDL, OpenVINO, Ray and Apache Spark https://github.com/intel-analytics/analytics-zoo AI on Big Data
Transformation of Big Data • Storing and processing more data • Analyzing (querying) more data • Real-time analysis • Modelling and prediction (ML/DL) AI is everywhere • Moving from experimentation to production • Applying to large-scale, distributed Big Data Big Data AI
Case Study: Image Feature Extraction at JD.com Query Search Result Source: “Bringing deep learning into big data analytics using BigDL”, Xianyan Jia and Zhenhua Wang, Strata Data Conference Singapore 2017 Similar Image Search Image Deduplication Image Feature Extraction: Applications:
* https://software.intel.com/en-us/articles/building-large-scale-image-feature-extraction-with-bigdl-at-jdcom * “BigDL: A Distributed Deep Learning Framework for Big Data”, ACM SoCC 2019, https://arxiv.org/abs/1804.05839 For more complete information about performance and benchmark results, visit www.intel.com/benchmarks. • End-to-end Big Data AI pipeline (using BigDL on Apache Spark) • Efficiently scale out (3.83x speed-up vs. Nvidia GPU severs)* Case Study: Image Feature Extraction at JD.com
Analytics Zoo: Software Platform for Big Data AI End-to-End Pipelines (Seamlessly scale AI models to distributed Big Data) ML Workflow (Automate tasks for building end-to-end pipelines) Models (Built-in models and algorithms) Compute Environment K8s Cluster Cloud Python Libraries (Numpy/Pandas/sklearn/…) DL Frameworks (TF/PyTorch/BigDL/OpenVINO/…) Distributed Analytics (Spark/Flink/Ray/…) Laptop Hadoop Cluster Powered by oneAPI https://github.com/intel-analytics/analytics-zoo
End-to-End Big Data Analytics and AI Seamless Scaling from Laptop to Distributed Big Data Big Data Pipeline Prototype on laptop using sample data Experiment on clusters with history data Production deployment w/ distributed data pipeline • Easily prototype end-to-end pipelines that apply AI models to big data • “Zero” code change from laptop to distributed cluster • Seamlessly deployed on production Hadoop/K8s clusters • Automate the process of applying machine learning to big data
Analytics Zoo: Software Platform for Big Data AI Recommendation Distributed TensorFlow & PyTorch on Spark Spark Dataframes & ML Pipelines for DL RayOnSpark InferenceModel Models & Algorithms End-to-end Pipelines Time Series Computer Vision NLP ML Workflow AutoML Automatic Cluster Serving Compute Environment K8s Cluster Cloud Python Libraries (Numpy/Pandas/sklearn/…) DL Frameworks (TF/PyTorch/BigDL/OpenVINO/…) Distributed Analytics (Spark/Flink/Ray/…) Laptop Hadoop Cluster Powered by oneAPI https://github.com/intel-analytics/analytics-zoo
Analytics Zoo: Software Platform for Big Data AI Recommendation Spark Dataframes & ML Pipelines for DL Distributed TensorFlow & PyTorch on Spark InferenceModel Models & Algorithms End-to-end Pipelines Time Series Computer Vision NLP ML Workflow AutoML Automatic Cluster Serving Compute Environment K8s Cluster Cloud Python Libraries (Numpy/Pandas/sklearn/…) DL Frameworks (TF/PyTorch/BigDL/OpenVINO/…) Distributed Analytics (Spark/Flink/Ray/…) Laptop Hadoop Cluster Powered by oneAPI RayOnSpark https://github.com/intel-analytics/analytics-zoo
Distributed TensorFlow/PyTorch on Spark Write TensorFlow/PyTorch inline with Spark code #pyspark code train_rdd = spark.hadoopFile(…).map(…) dataset = TFDataset.from_rdd(train_rdd,…) #tensorflow code import tensorflow as tf slim = tf.contrib.slim images, labels = dataset.tensors with slim.arg_scope(lenet.lenet_arg_scope()): logits, end_points = lenet.lenet(images, …) loss = tf.reduce_mean(tf.losses.sparse_softmax_cross_entropy( logits=logits, labels=labels)) #distributed training on Spark optimizer = TFOptimizer.from_loss(loss, Adam(…)) optimizer.optimize(end_trigger=MaxEpoch(5)) Analytics Zoo API in blue
Network Quality Prediction in SK Telecom Distributed TensorFlow/PyTorch on Spark https://networkbuilders.intel.com/solutionslibrary/sk-telecom-intel-build-ai-pipeline-to-improve-network-quality For more complete information about performance and benchmark results, visit www.intel.com/benchmarks.
• : distributed framework for emerging AI applications • RayOnSpark • Directly run Ray programs on Big Data cluster • Integrate Ray programs into Spark data pipeline https://medium.com/riselab/rayonspark-running-emerging-ai-applications-on-big-data-clusters-with-ray-and-analytics-zoo-923e0136ed6a RayOnSpark Run Ray Programs Directly on Big Data Platform
RayOnSpark Run Ray Programs Directly on Big Data Platform Analytics Zoo API in blue sc = init_spark_on_yarn(...) ray_ctx = RayContext(sc=sc, ...) ray_ctx.init() #Ray code @ray.remote class TestRay(): def hostname(self): import socket return socket.gethostname() actors = [TestRay.remote() for i in range(0, 100)] print([ray.get(actor.hostname.remote()) for actor in actors]) ray_ctx.stop() https://medium.com/riselab/rayonspark-running-emerging-ai-applications-on-big-data-clusters-with-ray-and-analytics-zoo-923e0136ed6a
Fast Food Recommendation in Burger King End-to-End Training Pipeline w/ RayOnSpark * https://medium.com/riselab/context-aware-fast-food-recommendation-at-burger-king-with-rayonspark-2e7a6009dd2d * “Context-Aware Drive-thru Recommendation Service at Fast Food Restaurants”, https://arxiv.org/abs/2010.06197 DATA INGESTIO N FEATURE ENGINEERING TRAINING INFERENCE on
Scalable AutoML for Time Series Prediction Automated feature generation, model selection and hyper parameter tuning Analytics Zoo API in blue tsp = TimeSequencePredictor( dt_col="datetime", target_col="value") pipeline = tsp.fit(train_df, val_df, metric="mse", recipe=RandomRecipe()) pipeline.predict(test_df) https://medium.com/riselab/scalable-automl-for-time-series-prediction-using-ray-and-analytics-zoo- b79a6fd08139
Scalable AutoML for Time Series Prediction Automated feature generation, model selection and hyper parameter tuning FeatureTransformer Model SearchEngine Search presets trial trial trial trial …best model /parameters trail jobs Pipeline with tunable parameters with tunable parameters configured with best parameters/model Each trial runs a different combination of hyper parameters Ray Tune rolling, scaling, feature generation, etc. Spark + Ray “Scalable AutoML for Time Series Forecasting using Ray”, USENIX OpML’20
TI-One ML Platform in Tencent Cloud Scalable AutoML for Time Series Prediction Using Analytics Zoo in Tencent Cloud TI- One ML Platform Predicting NYC Taxi Passengers Using AutoML https://software.intel.com/content/www/us/en/develop/articles/tencent-cloud-leverages-analytics-zoo-to-improve-performance-of-ti-one-ml-platform.html
“Zouwu” Open Source Framework for Time Series on Analytics Zoo Application framework for building end-to-end time series analysis • Use case - reference time series use cases • Models - built-in models for time series analysis • AutoTS - AutoML support for building E2E time series analysis pipelines Project Zouwu Built-in Models ML Workflow AutoML Workflow End-to-End Pipelines use-case models autots https://github.com/intel-analytics/analytics- zoo/tree/master/pyzoo/zoo/zouwu “Project Zouwu: Scalable AutoML for Telco Time Series Analysis using Ray and Analytics”, Ray Summit 2020
Analytics Zoo: Software Platform for Big Data AI • E2E Big Data & AI pipeline (distributed TF/PyTorch/OpenVINO/Ray on Spark) • Advanced AI workflow (AutoML, Time-Series, Cluster Serving, etc.) Github • Project repo: https://github.com/intel-analytics/analytics-zoo • Use cases: https://analytics-zoo.github.io/master/#powered-by/ Technical paper/tutorials • CVPR 2020 tutorial: https://jason-dai.github.io/cvpr2018/ • ACM SoCC 2019 paper: https://arxiv.org/abs/1804.05839 • AAAI 2019 tutorial: https://jason-dai.github.io/aaai2019/ • CVPR 2018 tutorial: https://jason-dai.github.io/cvpr2018/ Conclusion
Jason Dai Senior Principal Engineer
Analytics Zoo: Software Platform for Big Data AI End-to-End Pipelines (Seamlessly scale AI models to distributed Big Data) ML Workflow (Automate tasks for building end-to-end pipelines) Compute Environment K8s Cluster Cloud Python Libraries (Numpy/Pandas/sklearn/…) DL Frameworks (TF/PyTorch/BigDL/OpenVINO/…) Distributed Analytics (Spark/Flink/Ray/…) Laptop Hadoop Cluster Powered by oneAPI Recommendation Time Series Computer Vision NLP https://github.com/intel-analytics/analytics-zoo
• Recommendation • Time series analysis • Computer vision • Natural language processing (NLP) Big Data AI Applications on Analytics Zoo
Food Recommendation Using Analytics Zoo in Burger King Guest arrives ODMB Checks Menu Board Cashier enters order Checks Menu Board Guest completes order * https://medium.com/riselab/context-aware-fast-food-recommendation-at-burger-king-with-rayonspark-2e7a6009dd2d * “Context-aware Fast Food Recommendation with Ray on Apache Spark at Burger King”, Data + AI Summit Europe 2020
Food Recommendation Challenges Challenges • Lack of user identifiers • Same session food compatibilities • Other variables in our use case: locations, weathers, time, etc. • Deployment challenges * https://medium.com/riselab/context-aware-fast-food-recommendation-at-burger-king-with-rayonspark-2e7a6009dd2d * “Context-aware Fast Food Recommendation with Ray on Apache Spark at Burger King”, Data + AI Summit Europe 2020
Transformer Cross Transformer (TxT) Model Model Components • Sequence Transformer • Taking item order sequence as input • Context Transformer • Taking multiple context features as input • Latent Cross Joint Training • Element-wise product for both transformer outputs * https://medium.com/riselab/context-aware-fast-food-recommendation-at-burger-king-with-rayonspark-2e7a6009dd2d * “Context-aware Fast Food Recommendation with Ray on Apache Spark at Burger King”, Data + AI Summit Europe 2020
Unified Big Data Processing and Model Training on Analytics Zoo CurrentPrevious * https://medium.com/riselab/context-aware-fast-food-recommendation-at-burger-king-with-rayonspark-2e7a6009dd2d * “Context-aware Fast Food Recommendation with Ray on Apache Spark at Burger King”, Data + AI Summit Europe 2020
Food Recommendation Using Analytics Zoo in Burger King Offline Training Result Model Top1 Accuracy Top3 Accuracy RNN 29.98% 46.24% Contextual ItemCF 32.18% 48.37% RNN Latent Cross 33.10% 49.98% TxT 34.52% 52.37% A/B Testing Result Model Conversation Rate Gain Add-on Sales Gain RNN Latent Cross (control) - - TxT +7.5% +4.7% * https://medium.com/riselab/context-aware-fast-food-recommendation-at-burger-king-with-rayonspark-2e7a6009dd2d * “Context-aware Fast Food Recommendation with Ray on Apache Spark at Burger King”, Data + AI Summit Europe 2020
Recommendation Using Analytic Zoo in Mastercard https://software.intel.com/en-us/articles/deep-learning-with-analytic-zoo-optimizes-mastercard-recommender-ai-service Train NCF Model Features Models Model Candidates Models sampled partition Training Data … Load Parquet Train Multiple Models Train Wide & Deep Model sampled partition sampled partition Spark ML Pipeline Stages Test Data Predictions Test Spark DataFramesParquet Files Feature Selections SparkMLPipeline Neural Recommender using Analytics Zoo Estimator Transformer Model Evaluation & Fine Tune Train ALS Model
Time Series Based Network Quality Prediction in SK Telecom • Predict Network Quality Indicators (CQI, RSRP, RSRQ, SINR, …)* for anomaly detection and real-time management * CQI : Channel Quality Indicator * RSRP : Reference Signal Received Power * RSRQ : Reference Signal Received Quality * SINR :Signal to Interference Noise Ratio * “Vectorized Deep Learning Acceleration from Preprocessing to Inference and Training on Apache Spark in SK Telecom”, Spark + AI Summit 2020 * https://networkbuilders.intel.com/solutionslibrary/sk-telecom-intel-build-ai-pipeline-to-improve-network-quality
Memory Augmented Network https://networkbuilders.intel.com/solutionslibrary/sk-telecom-intel-build-ai-pipeline-to-improve-network-quality
Memory Augmented Network – Test Result Improved predictions for sudden change! seq2seq Mem-network https://networkbuilders.intel.com/solutionslibrary/sk-telecom-intel-build-ai-pipeline-to-improve-network-quality
Migrating to Analytics Zoo on Intel® Xeon Data Loader DRAM Store tiering forked. Flash Store customized. Data Source APIs Spark-SQL SQL Queries (Web, Jupyter) LegacyDesignwithGPU Export Preprocessing AITraining/Inference GPU Servers NewArchitecture: Unified DataAnalytic+AIPlatform Preprocessing RDDofTensor 2nd Generation Intel®Xeon® Scalable Processors csv files pandas, dask spark spark spark Manually manage separate clusters, and segregated workflow E2E architecture that atomically scales deep learning on Spark AIModelCodeofTF https://networkbuilders.intel.com/solutionslibrary/sk-telecom-intel-build-ai-pipeline-to-improve-network-quality
Inference Pipeline Speed-up with Analytics Zoo * https://networkbuilders.intel.com/solutionslibrary/sk-telecom-intel-build-ai-pipeline-to-improve-network-quality Up-to 6x speedup for end-to-end inference on Analytics Zoo in SK Telecom*
Training Pipeline Speed-up with Analytics Zoo Up-to 4x speedup for end-to-end training on Analytics Zoo in SK Telecom* * https://networkbuilders.intel.com/solutionslibrary/sk-telecom-intel-build-ai-pipeline-to-improve-network-quality
Wind Power Prediction using Analytics Zoo in GoldWind LSTNet ETL Training Prediction Deployment Update DB Historical Power Wind Power Prediction • Accuracy improved to 79% (from previous 59%)* • 4x training speedup* For more complete information about performance and benchmark results, visit www.intel.com/benchmarks. * https://www.intel.cn/content/www/cn/zh/analytics/artificial-intelligence/create-power-forecasting-solutions.html
Industrial Vision Inspection Using Analytics Zoo in Midea and KUKA https://software.intel.com/en-us/articles/industrial-inspection-platform-in-midea-and-kuka-using-distributed-tensorflow-on-analytics
Industrial Vision Inspection Using Analytics Zoo in Midea and KUKA Edge to Cloud architecture using Analytics Zoo • 99.8% accuracy* • <50ms image processing latency* • >8x inference speedup* * https://software.intel.com/en-us/articles/industrial-inspection-platform-in-midea-and-kuka-using-distributed-tensorflow-on-analytics * https://www.intel.cn/content/www/cn/zh/analytics/artificial-intelligence/midea-case-study.html For more complete information about performance and benchmark results, visit www.intel.com/benchmarks.
AI-Assisted Radiology Using Analytics Zoo in Dell EMC Condition E Condition D Condition C Condition B Condition A Patient A Transfer Learning using ResNet-50 trained with ImageNet https://www.delltaechnologies.com/resources/en-us/asset/white-papers/solutions/h17686_hornet_wp.pdf chest X-rays
Customer Service Chatbot Using Analytics Zoo in Microsoft Azure *https://software.intel.com/en-us/articles/use-analytics-zoo-to-inject-ai-into-customer-service-platforms-on-microsoft-azure-part-1 *https://www.infoq.com/articles/analytics-zoo-qa-module/
Job Recommendation Using Analytics Zoo in Talroo documents (resume, job description) https://software.intel.com/content/www/us/en/develop/articles/talroo-uses-analytics-zoo-and-aws-to-leverage-deep-learning-for-job-recommendations.html
Analytics Zoo: Software Platform for Big Data AI • E2E Big Data & AI pipeline (distributed TF/PyTorch/OpenVINO/Ray on Spark) • Advanced AI workflow (AutoML, Time-Series, Cluster Serving, etc.) Github • Project repo: https://github.com/intel-analytics/analytics-zoo • Use cases: https://analytics-zoo.github.io/master/#powered-by/ Technical paper/tutorials • CVPR 2020 tutorial: https://jason-dai.github.io/cvpr2018/ • ACM SoCC 2019 paper: https://arxiv.org/abs/1804.05839 • AAAI 2019 tutorial: https://jason-dai.github.io/aaai2019/ • CVPR 2018 tutorial: https://jason-dai.github.io/cvpr2018/ Big Data AI
Summary INDUSTRY INFLECTIONS ARE FUELING THE GROWTH OF DATA 5G Network Transformation, Artificial Intelligence, Intelligent Edge, Cloudification AI & ANALYTICS ARE THE DEFINING WORKLOADS OF THE NEXT DECADE with growing demand for end-to-end AI pipeline UNMATCHED PORTFOLIO BREADTH AND ECOSYSTEM SUPPORT Intel delivers a silicon & software foundation designed for the diverse range of use cases from the cloud to the edge ANALYTICS ZOO OPEN-SOURCE SOFTWARE PLATFORM FOR BIG DATA AI Simplifies End-to-End Big Data AI pipeline solutions development
Thank You
Notices & Disclaimers • Performance varies by use, configuration and other factors. Learn more at www.intel.com/performanceIndex • Performance may vary based on the specific game title and server configuration. To reference the full list of Intel Server GPU platform measurements, please refer to http://www.intel.com/content/www/us/en/benchmarks/server/graphics/IntelServerGPU • All product plans and roadmaps are subject to change without notice. • Intel technologies may require enabled hardware, software or service activation. • No product or component can be absolutely secure. • Your costs and results may vary. • Intel does not control or audit third-party data. You should consult other sources to evaluate accuracy. • All product plans and roadmaps are subject to change without notice. • © Intel Corporation. Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries. Other names and brands may be claimed as the property of others. Intel Server GPU TCO analysis is based on internal Intel research. Pricing as of 10/01/2020. Analysis assumes standard serving pricing, GPU list pricing, and software pricing based on estimated Nvidia software license costs of $1 per year for 5 years. • Intel Server GPU Performance may vary based on the specific game title and server configuration. To reference the full list of Intel Server GPU platform measurements, please refer to http://www.intel.com/content/www/us/en/benchmarks/server/graphics/IntelServerGPU • Video game footage courtesy of Tencent Games and Gamestream. • LEGO STAR WARS TITLES : © Lucasfilm Entertainment Company Ltd. or Lucasfilm Ltd. & ® or TM as indicated. All rights reserved. • LEGO, the LEGO logo and the Minifigure are trademarks of The LEGO Group. © The LEGO Group. All rights reserved. • “DiRT4”™ : © 2017 The Codemasters Software Company Limited ("Codemasters"). All rights reserved. "Codemasters"®, “EGO”®, the Codemasters logo, and “DiRT”® are registered trademarks owned by Codemasters. “DiRT4”™ and “RaceNet”™ are trademarks of Codemasters. All rights reserved. Under licence from International Management Group (UK) Limited. All other copyrights or trademarks are the property of their respective owners and are being used under license. Developed by Codemasters.

Check these out next

End-to-End Big Data AI with Analytics Zoo

Recommandé

Contenu connexe

Présentations pour vous(20)

Similaire à End-to-End Big Data AI with Analytics Zoo(20)

Dernier(20)

End-to-End Big Data AI with Analytics Zoo