Tesla Dojo is a series of supercomputers designed and built by Tesla for computer vision video processing and recognition. It was used for training Tesla's machine learning models to improve its Full Self-Driving (FSD) advanced driver-assistance system. It went into production in July 2023. Dojo's goal was to efficiently process millions of terabytes of video data captured from real-life driving situations from Tesla's 4+ million cars. This goal led to a considerably different architecture than conventional supercomputer designs. In August 2025, Bloomberg News reported that the Dojo project had been disbanded, though it was restarted in January 2026. == History == Tesla operates several massively parallel computing clusters for developing its Autopilot advanced driver assistance system. Its primary unnamed cluster using 5,760 Nvidia A100 graphics processing units (GPUs) was touted by Andrej Karpathy in 2021 at the fourth International Joint Conference on Computer Vision and Pattern Recognition (CCVPR 2021) to be "roughly the number five supercomputer in the world" at approximately 81.6 petaflops, based on scaling the performance of the Nvidia Selene supercomputer, which uses similar components. However, the performance of the primary Tesla GPU cluster has been disputed, as it was not clear if this was measured using single-precision or double-precision floating point numbers (FP32 or FP64). Tesla also operates a second 4,032 GPU cluster for training and a third 1,752 GPU cluster for automatic labeling of objects. The primary unnamed Tesla GPU cluster has been used for processing one million video clips, each ten seconds long, taken from Tesla Autopilot cameras operating in Tesla cars in the real world, running at 36 frames per second. Collectively, these video clips contained six billion object labels, with depth and velocity data; the total size of the data set was 1.5 petabytes. This data set was used for training a neural network intended to help Autopilot computers in Tesla cars understand roads. By August 2022, Tesla had upgraded the primary GPU cluster to 7,360 GPUs. Dojo was first mentioned by Elon Musk in April 2019 during Tesla's "Autonomy Investor Day". In August 2020, Musk stated it was "about a year away" due to power and thermal issues. Dojo was officially announced at Tesla's Artificial Intelligence (AI) Day on August 19, 2021. Tesla revealed details of the D1 chip and its plans for "Project Dojo", a datacenter that would house 3,000 D1 chips; the first "Training Tile" had been completed and delivered the week before. In October 2021, Tesla released a "Dojo Technology" whitepaper describing the Configurable Float8 (CFloat8) and Configurable Float16 (CFloat16) floating point formats and arithmetic operations as an extension of Institute of Electrical and Electronics Engineers (IEEE) standard 754. At the follow-up AI Day in September 2022, Tesla announced it had built several System Trays and one Cabinet. During a test, the company stated that Project Dojo drew 2.3 megawatts (MW) of power before tripping a local San Jose, California power substation. At the time, Tesla was assembling one Training Tile per day. In August 2023, Tesla powered on Dojo for production use as well as a new training cluster configured with 10,000 Nvidia H100 GPUs. In January 2024, Musk described Dojo as "a long shot worth taking because the payoff is potentially very high. But it's not something that is a high probability." In June 2024, Musk explained that ongoing construction work at Gigafactory Texas is for a computing cluster claiming that it is planned to comprise an even mix of "Tesla AI" and Nvidia/other hardware with a total thermal design power of at first 130 MW and eventually exceeding 500 MW. In August 2025, Bloomberg News reported that the Dojo project was disbanded, though Musk announced it would be restarted in January 2026 with a new chip iteration. == Technical architecture == The fundamental unit of the Dojo supercomputer is the D1 chip, designed by a team at Tesla led by ex-AMD CPU designer Ganesh Venkataramanan, including Emil Talpes, Debjit Das Sarma, Douglas Williams, Bill Chang, and Rajiv Kurian. The D1 chip is manufactured by the Taiwan Semiconductor Manufacturing Company (TSMC) using 7 nanometer (nm) semiconductor nodes, has 50 billion transistors and a large die size of 645 mm2 (1.0 square inch). Updating at Artificial Intelligence (AI) Day in 2022, Tesla announced that Dojo would scale by deploying multiple ExaPODs, in which there would be: 10 Cabinets per ExaPOD (1,062,000 cores, 3,000 D1 chips) 2 System Trays per Cabinet (106,200 cores, 300 D1 chips) 6 Training Tiles per System Tray (53,100 cores, along with host interface hardware) 25 D1 chips per Training Tile (8,850 cores) 354 computing cores per D1 chip According to Venkataramanan, Tesla's senior director of Autopilot hardware, Dojo will have more than an exaflop (a million teraflops) of computing power. For comparison, according to Nvidia, in August 2021, the (pre-Dojo) Tesla AI-training center used 720 nodes, each with eight Nvidia A100 Tensor Core GPUs for 5,760 GPUs in total, providing up to 1.8 exaflops of performance. === D1 chip === Each node (computing core) of the D1 processing chip is a general purpose 64-bit CPU with a superscalar core. It supports internal instruction-level parallelism, and includes simultaneous multithreading (SMT). It doesn't support virtual memory and uses limited memory protection mechanisms. Dojo software/applications manage chip resources. The D1 instruction set supports both 64-bit scalar and 64-byte single instruction, multiple data (SIMD) vector instructions. The integer unit mixes reduced instruction set computer (RISC-V) and custom instructions, supporting 8, 16, 32, or 64 bit integers. The custom vector math unit is optimized for machine learning kernels and supports multiple data formats, with a mix of precisions and numerical ranges, many of which are compiler composable. Up to 16 vector formats can be used simultaneously. ==== Node ==== Each D1 node uses a 32-byte fetch window holding up to eight instructions. These instructions are fed to an eight-wide decoder which supports two threads per cycle, followed by a four-wide, four-way SMT scalar scheduler that has two integer units, two address units, and one register file per thread. Vector instructions are passed further down the pipeline to a dedicated vector scheduler with two-way SMT, which feeds either a 64-byte SIMD unit or four 8×8×4 matrix multiplication units. The network on-chip (NOC) router links cores into a two-dimensional mesh network. It can send one packet in and one packet out in all four directions to/from each neighbor node, along with one 64-byte read and one 64-byte write to local SRAM per clock cycle. Hardware native operations transfer data, semaphores and barrier constraints across memories and CPUs. System-wide double data rate 4 (DDR4) synchronous dynamic random-access memory (SDRAM) memory works like bulk storage. ==== Memory ==== Each core has a 1.25 megabytes (MB) of SRAM main memory. Load and store speeds reach 400 gigabytes (GB) per second and 270 GB/sec, respectively. The chip has explicit core-to-core data transfer instructions. Each SRAM has a unique list parser that feeds a pair of decoders and a gather engine that feeds the vector register file, which together can directly transfer information across nodes. ==== Die ==== Twelve nodes (cores) are grouped into a local block. Nodes are arranged in an 18×20 array on a single die, of which 354 cores are available for applications. The die runs at 2 gigahertz (GHz) and totals 440 MB of SRAM (360 cores × 1.25 MB/core). It reaches 376 teraflops using 16-bit brain floating point (BF16) numbers or using configurable 8-bit floating point (CFloat8) numbers, which is a Tesla proposal, and 22 teraflops at FP32. Each die comprises 576 bi-directional serializer/deserializer (SerDes) channels along the perimeter to link to other dies, and moves 8 TB/sec across all four die edges. Each D1 chip has a thermal design power of approximately 400 watts. === Training Tile === The water-cooled Training Tile packages 25 D1 chips into a 5×5 array. Each tile supports 36 TB/sec of aggregate bandwidth via 40 input/output (I/O) chips - half the bandwidth of the chip mesh network. Each tile supports 10 TB/sec of on-tile bandwidth. Each tile has 11 GB of SRAM memory (25 D1 chips × 360 cores/D1 × 1.25 MB/core). Each tile achieves 9 petaflops at BF16/CFloat8 precision (25 D1 chips × 376 TFLOP/D1). Each tile consumes 15 kilowatts; 288 amperes at 52 volts. === System Tray === Six tiles are aggregated into a System Tray, which is integrated with a host interface. Each host interface includes 512 x86 cores, providing a Linux-based user environment. Previously, the Dojo System Tray was known as the Training Matrix, which includes six Training Tiles, 20 Dojo Interface Processor cards across four host servers, and Ethernet-l
Multisample anti-aliasing
Multisample anti-aliasing (MSAA) is a type of spatial anti-aliasing, a technique used in computer graphics to remove jaggies. It is an optimization of supersampling, where only the necessary parts are sampled more. Jaggies are only noticed in a small area, so the area is quickly found, and only that is anti-aliased. == Definition == The term generally refers to a special case of supersampling. Initial implementations of full-scene anti-aliasing (FSAA) worked conceptually by simply rendering a scene at a higher resolution, and then downsampling to a lower-resolution output. Most modern GPUs are capable of this form of anti-aliasing, but it greatly taxes resources such as texture, bandwidth, and fillrate. (If a program is highly TCL-bound or CPU-bound, supersampling can be used without much performance hit.) According to the OpenGL GL_ARB_multisample specification, "multisampling" refers to a specific optimization of supersampling. The specification dictates that the renderer evaluate the fragment program once per pixel, and only "truly" supersample the depth and stencil values. (This is not the same as supersampling but, by the OpenGL 1.5 specification, the definition had been updated to include fully supersampling implementations as well.) In graphics literature in general, "multisampling" refers to any special case of supersampling where some components of the final image are not fully supersampled. The lists below refer specifically to the ARB_multisample definition. == Description == In supersample anti-aliasing, multiple locations are sampled within every pixel, and each of those samples is fully rendered and combined with the others to produce the pixel that is ultimately displayed. This is computationally expensive, because the entire rendering process must be repeated for each sample location. It is also inefficient, as aliasing is typically only noticed in some parts of the image, such as the edges, whereas supersampling is performed for every single pixel. In multisample anti-aliasing, if any of the multi sample locations in a pixel is covered by the triangle being rendered, a shading computation must be performed for that triangle. However this calculation only needs to be performed once for the whole pixel regardless of how many sample positions are covered; the result of the shading calculation is simply applied to all of the relevant multi sample locations. In the case where only one triangle covers every multi sample location within the pixel, only one shading computation is performed, and these pixels are little more expensive than (and the result is no different from) the non-anti-aliased image. This is true of the middle of triangles, where aliasing is not an issue. (Edge detection can reduce this further by explicitly limiting the MSAA calculation to pixels whose samples involve multiple triangles, or triangles at multiple depths.) In the extreme case where each of the multi sample locations is covered by a different triangle, a different shading computation will be performed for each location and the results then combined to give the final pixel, and the result and computational expense are the same as in the equivalent supersampled image. The shading calculation is not the only operation that must be performed on a given pixel; multisampling implementations may variously sample other operations such as visibility at different sampling levels. == Advantages == The pixel shader usually only needs to be evaluated once per pixel for every triangle covering at least one sample point. The edges of polygons (the most obvious source of aliasing in 3D graphics) are anti-aliased. Since multiple subpixels per pixel are sampled, polygonal details smaller than one pixel that might have been missed without MSAA can be captured and made a part of the final rendered image if enough samples are taken. == Disadvantages == === Alpha testing === Alpha testing is a technique common to older video games used to render translucent objects by rejecting pixels from being written to the framebuffer. If the alpha value of a translucent fragment (pixel) is below a specified threshold, it will be discarded. Because this is performed on a pixel by pixel basis, the image does not receive the benefits of multi-sampling (all of the multisamples in a pixel are discarded based on the alpha test) for these pixels. The resulting image may contain aliasing along the edges of transparent objects or edges within textures, although the image quality will be no worse than it would be without any anti-aliasing. Translucent objects that are modelled using alpha-test textures will also be aliased due to alpha testing. This effect can be minimized by rendering objects with transparent textures multiple times, although this would result in a high performance reduction for scenes containing many transparent objects. === Aliasing === Because multi-sampling calculates interior polygon fragments only once per pixel, aliasing and other artifacts will still be visible inside rendered polygons where fragment shader output contains high frequency components. === Performance === While less performance-intensive than SSAA (supersampling), it is possible in certain scenarios (scenes heavy in complex fragments) for MSAA to be multiple times more intensive for a given frame than post processing anti-aliasing techniques such as FXAA, SMAA and MLAA. Early techniques in this category tend towards a lower performance impact, but suffer from accuracy problems. More recent post-processing based anti-aliasing techniques such as temporal anti-aliasing (TAA), which reduces aliasing by combining data from previously rendered frames, have seen the reversal of this trend, as post-processing AA becomes both more versatile and more expensive than MSAA, which cannot antialias an entire frame alone. == Sampling methods == === Point sampling === In a point-sampled mask, the coverage bit for each multisample is only set if the multisample is located inside the rendered primitive. Samples are never taken from outside a rendered primitive, so images produced using point-sampling will be geometrically correct, but filtering quality may be low because the proportion of bits set in the pixel's coverage mask may not be equal to the proportion of the pixel that is actually covered by the fragment in question. === Area sampling === Filtering quality can be improved by using area sampled masks. In this method, the number of bits set in a coverage mask for a pixel should be proportionate to the actual area coverage of the fragment. This will result in some coverage bits being set for multisamples that are not actually located within the rendered primitive, and can cause aliasing and other artifacts. == Sample patterns == === Regular grid === A regular grid sample pattern, where multisample locations form an evenly spaced grid throughout the pixel, is easy to implement and simplifies attribute evaluation (i.e. setting subpixel masks, sampling color and depth). This method is computationally expensive due to the large number of samples. Edge optimization is poor for screen-aligned edges, but image quality is good when the number of multisamples is large. === Sparse regular grid === A sparse regular grid sample pattern is a subset of samples that are chosen from the regular grid sample pattern. As with the regular grid, attribute evaluation is simplified due to regular spacing. The method is less computationally expensive due to having a fewer samples. Edge optimization is good for screen aligned edges, and image quality is good for a moderate number of multisamples. === Stochastic sample patterns === A stochastic sample pattern is a random distribution of multisamples throughout the pixel. The irregular spacing of samples makes attribute evaluation complicated. The method is cost efficient due to low sample count (compared to regular grid patterns). Edge optimization with this method, although sub-optimal for screen aligned edges. Image quality is excellent for a moderate number of samples. == Quality == Compared to supersampling, multisample anti-aliasing can provide similar quality at higher performance, or better quality for the same performance. Further improved results can be achieved by using rotated grid subpixel masks. The additional bandwidth required by multi-sampling is reasonably low if Z and colour compression are available. Most modern GPUs support 2×, 4×, and 8× MSAA samples. Higher values result in better quality, but are slower.
TimeTiger
TimeTiger is a time and project tracking app developed by Indigo Technologies Ltd. in Toronto, Ontario, Canada. Indigo was founded in 1997 and initially released TimeTiger in 1998. == Company == The company was incorporated in 1997 and began operations as a custom software developer. TimeTiger (internally called TaskMaster) was developed as a tool to help with Indigo's own project planning and estimating. After releasing TimeTiger as a commercial product in 1998, Indigo shifted its focus to time and project management solutions. TimeTiger first introduced support for web-based time logging in 2000, to appeal to workers who were not already tracking their time for billing reasons. Subsequent development emphasized project analysis tools. == Features == Web-based electronic time log "To Do" list to monitor project and non-project activities Pivot table report designer Role-based access control == Software integration == Reports can be exported to Microsoft Excel or saved as Excel-compatible HTML files. Microsoft Project files can be imported and exported. A Software Development Kit is available.
View model
A view model or viewpoints framework in systems engineering, software engineering, and enterprise engineering is a framework which defines a coherent set of views to be used in the construction of a system architecture, software architecture, or enterprise architecture. A view is a representation of the whole system from the perspective of a related set of concerns. Since the early 1990s there have been a number of efforts to prescribe approaches for describing and analyzing system architectures. A result of these efforts have been to define a set of views (or viewpoints). They are sometimes referred to as architecture frameworks or enterprise architecture frameworks, but are usually called "view models". Usually a view is a work product that presents specific architecture data for a given system. However, the same term is sometimes used to refer to a view definition, including the particular viewpoint and the corresponding guidance that defines each concrete view. The term view model is related to view definitions. == Overview == The purpose of views and viewpoints is to enable humans to comprehend very complex systems, to organize the elements of the problem and the solution around domains of expertise and to separate concerns. In the engineering of physically intensive systems, viewpoints often correspond to capabilities and responsibilities within the engineering organization. Most complex system specifications are so extensive that no single individual can fully comprehend all aspects of the specifications. Furthermore, we all have different interests in a given system and different reasons for examining the system's specifications. A business executive will ask different questions of a system make-up than would a system implementer. The concept of viewpoints framework, therefore, is to provide separate viewpoints into the specification of a given complex system in order to facilitate communication with the stakeholders. Each viewpoint satisfies an audience with interest in a particular set of aspects of the system. Each viewpoint may use a specific viewpoint language that optimizes the vocabulary and presentation for the audience of that viewpoint. Viewpoint modeling has become an effective approach for dealing with the inherent complexity of large distributed systems. Architecture description practices, as described in IEEE Std 1471-2000, utilize multiple views to address several areas of concerns, each one focusing on a specific aspect of the system. Examples of architecture frameworks using multiple views include Kruchten's "4+1" view model, the Zachman Framework, TOGAF, DoDAF, and RM-ODP. == History == In the 1970s, methods began to appear in software engineering for modeling with multiple views. Douglas T. Ross and K.E. Schoman in 1977 introduce the constructs context, viewpoint, and vantage point to organize the modeling process in systems requirements definition. According to Ross and Schoman, a viewpoint "makes clear what aspects are considered relevant to achieving ... the overall purpose [of the model]" and determines How do we look at [a subject being modelled]? As examples of viewpoints, the paper offers: Technical, Operational and Economic viewpoints. In 1992, Anthony Finkelstein and others published a very important paper on viewpoints. In that work: "A viewpoint can be thought of as a combination of the idea of an “actor”, “knowledge source”, “role” or “agent” in the development process and the idea of a “view” or “perspective” which an actor maintains." An important idea in this paper was to distinguish "a representation style, the scheme and notation by which the viewpoint expresses what it can see" and "a specification, the statements expressed in the viewpoint's style describing particular domains". Subsequent work, such as IEEE 1471, preserved this distinction by utilizing two separate terms: viewpoint and view, respectively. Since the early 1990s there have been a number of efforts to codify approaches for describing and analyzing system architectures. These are often termed architecture frameworks or sometimes viewpoint sets. Many of these have been funded by the United States Department of Defense, but some have sprung from international or national efforts in ISO or the IEEE. Among these, the IEEE Recommended Practice for Architectural Description of Software-Intensive Systems (IEEE Std 1471-2000) established useful definitions of view, viewpoint, stakeholder and concern and guidelines for documenting a system architecture through the use of multiple views by applying viewpoints to address stakeholder concerns. The advantage of multiple views is that hidden requirements and stakeholder disagreements can be discovered more readily. However, studies show that in practice, the added complexity of reconciling multiple views can undermine this advantage. IEEE 1471 (now ISO/IEC/IEEE 42010:2011, Systems and software engineering — Architecture description) prescribes the contents of architecture descriptions and describes their creation and use under a number of scenarios, including precedented and unprecedented design, evolutionary design, and capture of design of existing systems. In all of these scenarios the overall process is the same: identify stakeholders, elicit concerns, identify a set of viewpoints to be used, and then apply these viewpoint specifications to develop the set of views relevant to the system of interest. Rather than define a particular set of viewpoints, the standard provides uniform mechanisms and requirements for architects and organizations to define their own viewpoints. In 1996 the ISO Reference Model for Open Distributed Processing (RM-ODP) was published to provide a useful framework for describing the architecture and design of large-scale distributed systems. == View model topics == === View === A view of a system is a representation of the system from the perspective of a viewpoint. This viewpoint on a system involves a perspective focusing on specific concerns regarding the system, which suppresses details to provide a simplified model having only those elements related to the concerns of the viewpoint. For example, a security viewpoint focuses on security concerns and a security viewpoint model contains those elements that are related to security from a more general model of a system. A view allows a user to examine a portion of a particular interest area. For example, an Information View may present all functions, organizations, technology, etc. that use a particular piece of information, while the Organizational View may present all functions, technology, and information of concern to a particular organization. In the Zachman Framework views comprise a group of work products whose development requires a particular analytical and technical expertise because they focus on either the “what,” “how,” “who,” “where,” “when,” or “why” of the enterprise. For example, Functional View work products answer the question “how is the mission carried out?” They are most easily developed by experts in functional decomposition using process and activity modeling. They show the enterprise from the point of view of functions. They also may show organizational and information components, but only as they relate to functions. === Viewpoints === In systems engineering, a viewpoint is a partitioning or restriction of concerns in a system. Adoption of a viewpoint is usable so that issues in those aspects can be addressed separately. A good selection of viewpoints also partitions the design of the system into specific areas of expertise. Viewpoints provide the conventions, rules, and languages for constructing, presenting and analysing views. In ISO/IEC 42010:2007 (IEEE-Std-1471-2000) a viewpoint is a specification for an individual view. A view is a representation of a whole system from the perspective of a viewpoint. A view may consist of one or more architectural models. Each such architectural model is developed using the methods established by its associated architectural system, as well as for the system as a whole. === Modeling perspectives === Modeling perspectives is a set of different ways to represent pre-selected aspects of a system. Each perspective has a different focus, conceptualization, dedication and visualization of what the model is representing. In information systems, the traditional way to divide modeling perspectives is to distinguish the structural, functional and behavioral/processual perspectives. This together with rule, object, communication and actor and role perspectives is one way of classifying modeling approaches === Viewpoint model === In any given viewpoint, it is possible to make a model of the system that contains only the objects that are visible from that viewpoint, but also captures all of the objects, relationships and constraints that are present in the system and relevant to that viewpoint. Such a model is said to be a viewpoint model, or a view of the
EnQuire
Enquire is a web-based software application used as a platform for project, contract and grant management, as well as reporting and planning. Initially designed for the specific business requirements of the Australian Government, Queensland Government and Queensland Regional Bodies to manage natural resource projects, Enquire has since seen adoption outside of this industry and user segment. The use of Enquire by Natural Resource Management bodies within Queensland has been cited as a reason for the improved efficiency, quantity and quality of reporting. Technically, Enquire is implemented as a Java application built on a MySQL database. Enquire is hosted and supported under the software as a service model by Tactiv Pty Ltd. == History == The system was first released in 2005 under the name ViSTA NRM Online, proactively changing its name to Enquire in 2007 to avoid possible confusion with Windows Vista, which was being released at the time. In 2012, the Enquire project and support team was commercialized as its own company called Tactiv Pty Ltd. Tactiv is based predominantly in Brisbane, Australia. Tactiv has continued to develop and grow the Enquire Grant, Contract and Project management solution, releasing a new platform in 2017. Since commercialization, Tactiv has grown its client base to include government and non-government organizations such as foundations and not-for-profit organizations. == Functionality == The functionality of Enquire can be broken down into 5 key lifecycle solutions, all fully integrated and supported by over 40 feature rich and configurable modules: Grant Management Contract Management Project Portfolio Management Procurement Management Relationship Management The system provides its platform to meet the needs of "off the shelf" customers looking for a ready to use best practice option as well as a fully configurable option for specific requirements. The system offers a client supplier portal for external applicants or suppliers, a management portal for internal team usage and an administration portal for clients to manage access, roles, information, and other configurations. Key functional modules include: Online authoring and publishing for forms and applications Workflows Project Tracking Performance Reporting Financial Reporting Stakeholder Communication Budget management Document Management Milestone tracking Payments and Variations Management KPI tracking and Impact reporting The Enquire system is used to report against the Queensland Government's Q2 Coast and Country Program and parts of the Australian Government's Caring for our Country program. There is also a strategic planning module, which provides functionality to manage core-business administration and reporting requirements, whilst providing visibility of key activities and their alignment against organizational goals and strategic objectives. The systems architecture supports a range of implementation models with the capacity to manage one-to-one, one-to-many and many-to-many relationships between investors and investees. Under the usage model within Queensland, Regional Bodies use Enquire to load project contracts and report against these online. The regional bodies also record output, target and financial information in Enquire, which can then be used for operational purposes including financial, performance and target reporting. == External Audit == The Australian National Audit Office Audit Report No.21 2007–08 undertook a case study on Enquire. It noted: "The Queensland Department of Environment and Resource Management has developed the first integrated web-based system [Enquire] to manage performance information about Natural Resource Management activities in Queensland." Four of Queensland's 14 regional bodies commented on Enquire through the ANAO's survey. These four regional bodies indicated that Enquire offers a means of consistent reporting at the State level.
Convolutional layer
In artificial neural networks, a convolutional layer is a type of network layer that applies a convolution operation to the input. Convolutional layers are some of the primary building blocks of convolutional neural networks (CNNs), a class of neural network most commonly applied to images, video, audio, and other data that have the property of uniform translational symmetry. The convolution operation in a convolutional layer involves sliding a small window (called a kernel or filter) across the input data and computing the dot product between the values in the kernel and the input at each position. This process creates a feature map that represents detected features in the input. == Concepts == === Kernel === Kernels, also known as filters, are small matrices of weights that are learned during the training process. Each kernel is responsible for detecting a specific feature in the input data. The size of the kernel is a hyperparameter that affects the network's behavior. === Convolution === For a 2D input x {\displaystyle x} and a 2D kernel w {\displaystyle w} , the 2D convolution operation can be expressed as: y [ i , j ] = ∑ m = 0 k h − 1 ∑ n = 0 k w − 1 x [ i + m , j + n ] ⋅ w [ m , n ] {\displaystyle y[i,j]=\sum _{m=0}^{k_{h}-1}\sum _{n=0}^{k_{w}-1}x[i+m,j+n]\cdot w[m,n]} where k h {\displaystyle k_{h}} and k w {\displaystyle k_{w}} are the height and width of the kernel, respectively. This generalizes immediately to nD convolutions. Commonly used convolutions are 1D (for audio and text), 2D (for images), and 3D (for spatial objects, and videos). === Stride === Stride determines how the kernel moves across the input data. A stride of 1 means the kernel shifts by one pixel at a time, while a larger stride (e.g., 2 or 3) results in less overlap between convolutions and produces smaller output feature maps. === Padding === Padding involves adding extra pixels around the edges of the input data. It serves two main purposes: Preserving spatial dimensions: Without padding, each convolution reduces the size of the feature map. Handling border pixels: Padding ensures that border pixels are given equal importance in the convolution process. Common padding strategies include: No padding/valid padding. This strategy typically causes the output to shrink. Same padding: Any method that ensures the output size same as input size is a same padding strategy. Full padding: Any method that ensures each input entry is convolved over for the same number of times is a full padding strategy. Common padding algorithms include: Zero padding: Add zero entries to the borders of input. Mirror/reflect/symmetric padding: Reflect the input array on the border. Circular padding: Cycle the input array back to the opposite border, like a torus. The exact numbers used in convolutions is complicated, for which we refer to (Dumoulin and Visin, 2018) for details. == Variants == === Standard === The basic form of convolution as described above, where each kernel is applied to the entire input volume. === Depthwise separable === Depthwise separable convolution separates the standard convolution into two steps: depthwise convolution and pointwise convolution. The depthwise separable convolution decomposes a single standard convolution into two convolutions: a depthwise convolution that filters each input channel independently and a pointwise convolution ( 1 × 1 {\displaystyle 1\times 1} convolution) that combines the outputs of the depthwise convolution. This factorization significantly reduces computational cost. It was first developed by Laurent Sifre during an internship at Google Brain in 2013 as an architectural variation on AlexNet to improve convergence speed and model size. === Dilated === Dilated convolution, or atrous convolution, introduces gaps between kernel elements, allowing the network to capture a larger receptive field without increasing the kernel size. === Transposed === Transposed convolution, also known as deconvolution, fractionally strided convolution, and upsampling convolution, is a convolution where the output tensor is larger than its input tensor. It's often used in encoder-decoder architectures for upsampling. It's used in image generation, semantic segmentation, and super-resolution tasks. == History == The concept of convolution in neural networks was inspired by the visual cortex in biological brains. Early work by Hubel and Wiesel in the 1960s on the cat's visual system laid the groundwork for artificial convolution networks. An early convolution neural network was developed by Kunihiko Fukushima in 1969. It had mostly hand-designed kernels inspired by convolutions in mammalian vision. In 1979 he improved it to the Neocognitron, which learns all convolutional kernels by unsupervised learning (in his terminology, "self-organized by 'learning without a teacher'"). During the 1988 to 1998 period, a series of CNN were introduced by Yann LeCun et al., ending with LeNet-5 in 1998. It was an early influential CNN architecture for handwritten digit recognition, trained on the MNIST dataset, and was used in ATM. (Olshausen & Field, 1996) discovered that simple cells in the mammalian primary visual cortex implement localized, oriented, bandpass receptive fields, which could be recreated by fitting sparse linear codes for natural scenes. This was later found to also occur in the lowest-level kernels of trained CNNs. The field saw a resurgence in the 2010s with the development of deeper architectures and the availability of large datasets and powerful GPUs. AlexNet, developed by Alex Krizhevsky et al. in 2012, was a catalytic event in modern deep learning. In that year’s ImageNet competition, the AlexNet model achieved a 16% top-five error rate, significantly outperforming the next best entry, which had a 26% error rate. The network used eight trainable layers, approximately 650,000 neurons, and around 60 million parameters, highlighting the impact of deeper architectures and GPU acceleration on image recognition performance. From the 2013 ImageNet competition, most entries adopted deep convolutional neural networks, building on the success of AlexNet. Over the following years, performance steadily improved, with the top-five error rate falling from 16% in 2012 and 12% in 2013 to below 3% by 2017, as networks grew increasingly deep.
Cloud Native Computing Foundation
The Cloud Native Computing Foundation (CNCF) is a subsidiary of the Linux Foundation founded in 2015 to support cloud-native computing. == History == It was announced alongside Kubernetes 1.0, an open source container cluster manager, which was contributed to the Linux Foundation by Google as a seed technology. Founding members include Google, CoreOS, Mesosphere, Red Hat, Twitter, Huawei, Intel, RX-M, Cisco, IBM, Docker, Univa, and VMware. Today, CNCF is supported by over 450 members. In August 2018 Google announced that it was handing over operational control of Kubernetes to the community. == Projects == Argo is a collection of tools for getting work done with Kubernetes. Among its main features are Workflows and Events. It was accepted to CNCF on March 26, 2020 at the Incubating maturity level and then moved to the Graduated maturity level on December 6, 2022. cert-manager provisions and manages TLS certificates in Kubernetes. It was accepted to CNCF on November 10, 2020, moved to the Incubating maturity level on September 19, 2022, and then moved to the Graduated maturity level on September 29, 2024. Cilium provides networking, security, and observability for Kubernetes deployments using eBPF technology. It joined the CNCF at incubation level in October 2021 and the CNCF announced its graduation in October 2023. containerd is an industry-standard core container runtime. It is currently available as a daemon for Linux and Windows, which can manage the complete container lifecycle of its host system. In 2015, Docker donated the OCI Specification to The Linux Foundation with a reference implementation called runc. Since February 28, 2019 it is an official CNCF project. Its general availability and intention to donate the project to CNCF was announced by Docker in 2017. CoreDNS is a DNS server that chains plugins. Its graduation was announced in 2019. Dapr, the distributed application runtime, provides APIs for building secure and reliable microservices and agentic AI systems. Dapr was donated to the CNCF in November 2021 and joined at incubation level. The CNCF announced its graduation in November 2024. Envoy: Originally built at Lyft to move their architecture away from a monolith, Envoy is a high-performance open source edge and service proxy that makes the network transparent to applications. Lyft contributed Envoy to Cloud Native Computing Foundation in September 2017. etcd is a distributed key value store, providing a method of storing data across a cluster of machines. It became a CNCF incubating project in 2018 at KubeCon+CloudNativeCon North America in Seattle that year. Falco is an open source and cloud native runtime security initiative. It is the "de facto Kubernetes threat detection engine". It became an incubating project in January 2020 and graduated in February 2024. Flux is an open source project for powering GitOps in Kubernetes clusters. It provides the GitOps Toolkit, a set of Kubernetes APIs that allow you to define how configuration source code is securely pulled into your cluster and deployed by popular Kubernetes manifests rendering engines like Kustomize and Helm. The most recommended source mechanism is the OCIRepository API, which provides enhanced security and benefits from container image tooling out there. Flux has also notification integrations with popular services like Prometheus Alertmanager, PagerDuty, Slack and so on. Flux has graduated in CNCF in 2022. Harbor is an "open source trusted cloud native registry project that stores, signs, and scans content." It became an incubating project in September 2019 and graduated in June 2020. Helm is a package manager that helps developers "easily manage and deploy applications onto the Kubernetes cluster." It joined the incubating level in June 2018 and graduated in April 2020. Istio is a service mesh technology. It was accepted by CNCF in September 2022 and graduated on July 12, 2023. Jaeger, Created by Uber Engineering, Jaeger is an open source distributed tracing system inspired by Google Dapper paper and OpenZipkin community. It can be used for tracing microservice-based architectures, including distributed context propagation, distributed transaction monitoring, root cause analysis, service dependency analysis, and performance/latency optimization. The Cloud Native Computing Foundation Technical Oversight Committee voted to accept Jaeger as the 12th hosted project in September 2017 and became a graduated project in 2019. In 2020 it became an approved and fully integrated part of the CNCF ecosystem. Kubernetes is an open source framework for automating deployment and managing applications in a containerized and clustered environment. "It aims to provide better ways of managing related, distributed components across the varied infrastructure." It was originally designed by Google and donated to The Linux Foundation to form the Cloud Native Computing Foundation with Kubernetes as the seed technology. The "large and diverse" community supporting the project has made its staying power more robust than other, older technologies of the same ilk. In January 2020, the CNCF annual report showed significant growth in interest, training, event attendance and investment related to Kubernetes. Linkerd is CNCF's fifth member project, and the project that coined the term "service mesh". Linkerd adds observability, security, and reliability features to applications by adding them to the platform rather than the application layer, and features a "micro-proxy" to maximize speed and security of its data plane. Linkerd graduated from CNCF in July 2021. Open Policy Agent (OPA) is "an open source general-purpose policy engine and language for cloud infrastructure." It became a CNCF incubating project in April 2019. OPA graduated from CNCF in February 2021. Prometheus is a cloud monitoring tool sponsored by SoundCloud in early iterations. In August 2018, the tool was designated a graduated project by the Cloud Native Computing Foundation. It is now a Cloud Native Computing Foundation member project. Rook is CNCF's first cloud native storage project. It became an incubation level project in 2018 and graduated in October 2020. SPIFFE is an open standard and framework for workload identity, much the same way that OAuth is an open standard and framework for human identity. It is built from the ground up to accommodate modern computing environments, which operate with systems scale and velocity (as opposed to human scale and velocity), while still maintaining interoperability with existing technologies like OAuth and X.509 Public key infrastructure. Unlike other identity standards, SPIFFE supports multiple credential types for a single identity, ensuring that the highly varied needs of production environments are consistently met without compromise. SPIFFE joined the CNCF as a sandbox project in 2018, was accepted to incubation in 2020, and graduated in 2022. SPIRE is an open source identity provider for workloads based on the SPIFFE framework. It is highly pluggable, and fills the attestation and issuance needs required by any workload identity solution. The plugin interfaces it exposes allows users to write integrations with in-house systems, build internal self-service portals, and more. It is a very powerful building block for issuing short-lived identity credentials to dynamic cloud workloads. SPIRE became a CNCF Graduated project in 2022. The Update Framework (TUF) helps developers to secure new or existing software update systems, which are often found to be vulnerable to many known attacks. TUF addresses this widespread problem by providing a comprehensive, flexible security framework that developers can integrate with any software update system. TUF was CNCF's first security-focused project and the ninth project overall to graduate from the foundation's hosting program. TiKV provides a distributed key–value database. Vitess is a database clustering system for horizontal scaling of MySQL, first created for internal use by YouTube. It became a CNCF project in 2018 and graduated in November 2019. Contour is a management server for Envoy that can direct the management of Kubernetes' traffic. Contour also provides routing features that are more advanced than Kubernetes' out-of-the-box Ingress specification. VMWare contributed the project to CNCF in July 2020. Cortex offers horizontally scalable, multi-tenant, long-term storage for Prometheus and works alongside Amazon DynamoDB, Google Bigtable, Cassandra, S3, GCS, and Microsoft Azure. It was introduced into the ecosystem incubator alongside Thanos in August 2020. CRI-O is an Open Container Initiative (OCI) based "implementation of Kubernetes Container Runtime Interface". CRI-O allows Kubernetes to be container runtime-agnostic. It became an incubating project in 2019. gRPC is a "modern open source high performance RPC framework that can run in any environment." The project was formed in 2015 when Google decided to open sou