China
Africa
Explore chapters and articles related to this topic
Storage and databases for big data
Published in Jun Deng, Lei Xing, Big Data in Radiation Oncology, 2019
Tomas Skripcak, Uwe Just, Ida Schönfeld, Esther G.C. Troost, Mechthild Krause
Key–value stores are based on the idea of distributed hash tables searchable by a key that identifies the corresponding value (binary data), see Figure 3.4a. In computer science, these sorts of data structures are also known as associative arrays. Intrinsic operations that are supported by a key–value store itself are limited to the management of key–value pairs such as insert new pair, reassign value, remove pair, or lookup according to a key. Any other data processing (including reference integrity checking) has to be performed on the client site by the power-user who has the knowledge about the internal structure of values within the key–value store and the key naming conventions. Key–value store technology is often used as a low-level modeling scheme even for more complex data models. Different implementations of key–value stores can provide extended support for specialized value types declared as lists and sets or documents using Extensible Markup Language (XML) or JavaScript Object Notation (JSON) encoding. Often they use an in-memory storage mechanism but can also operate on a distributed file system.
Server-Side Technologies
Published in Akshi Kumar, Web Technology, 2018
Arrays are complex variables that allow us to store more than one value or a group of values under a single variable name. There are three types of arrays that you can create in PHP. These are: Indexed array: An array with a numeric key. Here the indices of the data elements are numbers that start with 0 (zero indexed) and grow incrementally. Here the indexes of the 3 elements or values are 0, 1, and 2, respectively. For example: This is equivalent to the following example, in which indexes are assigned manually: Associative array: An array where each key has its own specific value. That is, in an associative array, the keys assigned to values can be arbitrary and user defined strings. In the following example, the array uses keys instead of index numbers: Multidimensional array: An array containing one or more arrays within it. In a multidimensional array, each element can also be an array and each element in the sub-array can be an array or further contain an array within itself and so on. An example of a multidimensional array will look something like this Figure 9.8.
Grid k-d tree approach for point location in polyhedral data sets – application to explicit MPC
Published in International Journal of Control, 2020
In addition to the BST data structure mentioned above, other data structures have been applied to the point location problem. The hash table is a data structure used to implement an associative array where the time complexity is O(1). A two-stage algorithm was proposed which combines the direct search method with the hash table (Bayat, Johansen, & Jalali, 2011). In this approach, the hash table divides the whole state space into many grids in which the number of affine control laws differs greatly, it then locates the corresponding partition using the direct search method, which produces a fairly low online computation efficiency. To improve online searching efficiency a two-level hash table structure and grid-BST structure based on the hash table method have been used. (Zhang, Xiu, Xie, & Hu, 2016). Other data structures are also used to solve the point location problem in EMPC, namely, quardtree, graph, connected graph, etc. (Herceg, Jones, Kvasnica, & Morari, 2015; Herceg, Mariethoz, & Morari, 2013; Jafargholi, Peyrl, Zanarini, & Herceg, 2014; Oberdieck, Diangelakis, & Pistikopoulos, 2017).
A quantized approach for occupancy grids for autonomous vehicles: Q-Trees
Published in Advanced Robotics, 2018
The pseudocode in Algorithm 1 explains the assignment procedure for obstacles. This method assigns the landmarks in the ROI to the left or right child through the graph tree branches. Every node has the value of the sampling rate of its own. These landmarks are stored in an associative array to list the occluded nodes and which landmarks are assigned to a specific node. The time complexity of this method takes steps in all cases (best, worst and the average). Variable m represents the depth of the farthest leaves in the branch. Therefore, the ultimate worst scenario takes and this term is defined as so, the worst case for searching the dependent node of an obstacle takes
Improving dynamic programming for travelling salesman with precedence constraints: parallel Morin–Marsten bounding
Published in Optimization Methods and Software, 2021
Yaroslav. V. Salii, Andrey S. Sheka
Later, in [46], this data structure was amended as follows: std::vector<std::unordered_map<std::bitset, std::map<uint16_t, uint32_t> > >, notably, (a) order ideals were encoded with std::bitset instead of 32-bit unsigned integers uint32_t, to accommodate the TSPLIB instances, which have as much as 378 cities (rbg378a.sop); (b) per the TSPLIB custom, the travel costs were integer, not floating-point; (c) associative array std::map was used instead of the hash table for the innermost container, in an attempt to provide more consistent memory footprint.