Aventuras bushveldNao ha nada como o mato africano e nao ha melhor sitio no mundo para caça grossa. Alem dos seus famosos parques nacionais, a África do Sul tem uma serie de reservas privadas, todas elas proporcionando excelente vida selvagem e ornitologia.Ou junte-se a uma excursao via terrestre ou um safari guiado, onde pode recostar-se e concentrar a sua atençao a tentar vislumbrar os animais de grande porte enquanto alguem conduz por si.Para uma experiencia de caça absolutamente unica, precisara de visitar o territorio lowveld das provincias do Limpopo, Mpumalanga, Nororeste ou KwaZulu-Natal, onde os elefantes deslizam graciosamente por entre o mato e os leoes descansam na hora mais quente do dia depois de uma noite de caça.Na provincia de Gauteng, a pouco mais de uma hora de carro das selvas urbanas de Joanesburgo e Pretoria, pode observar leoes, elefantes, bufalos e centenas de outras especies no seu habitat natural.No Cabo Ocidental, com o seu clima e vegetaçao diferentes, nao ira encontrar leoes nem elefantes, mas podera ver as gazelas tipicas da África do Sul, as springbok, as zebras-do-cabo, os bontebok ou antilopes, os black wildebeest ou gnus, e tantos outros.O Cabo Oriental e uma zona de transiçao entre o Cabo Ocidental e as zonas de caça do Lowveld. A provincia esta rapidamente a tornar-se numa zona de safari muito procurada, sem duvida por nao ter indicios de malaria. O Parque Nacional Addo Elefante esta constantemente a crescer e estender-se-a por uma enorme variedade de comunidades ecologicas distintas, desde as marinhas as de montanha. Existem igualmente algumas reservas privadas fantasticas nesta mesma provincia, de realçar Shamwari.O Parque Nacional Golden Gate, na provincia do Estado Livre (Free State), e famoso pela caça de alta altitude, como o antilope africano (eland) e o gnu (black wildebeest).O Cabo Norte e muito arido e um gosto adquirido, ha que se aprender a gostar, mas mesmo assim possui destinos de caça excelentes. O Parque Nacional de Augrabies Falls e basicamente paisagem, embora tenha uma admiravel vida animal e de aves. O Parque Transfronteiriço Kgalagadi, o primeiro parque entre fronteiras, e famoso pelos seus enormes leoes Kalahari com juba preta e o elegante orix (gemsbok), que ali sao encontrados em abundancia.BiodiversidadeA África do Sul possui o terceiro maior nivel de biodiversidade do mundo, englobando sete dos maiores tipos de habitat terrestres, ou zonas de vida ecologica, com condiçoes ambientais distintas e respectiva fauna e flora.Cerca de 10% das especies de flores de todo o mundo pode ser encontrada na África do Sul e e o unico pais do mundo que detem, dentro das suas fronteiras, um verdadeiro reino vegetal.A Regiao Floral do Cabo, um dos oito Patrimonios Mundiais da África do Sul, abrange oito areas protegidas, que vao desde a Peninsula do Cabo ao Cabo Oriental, desbravando um cenario deslumbrante de montanha e de oceano, contendo alguma da biodiversidade floristica mais rica do mundo.O Parque iSimangaliso Wetland, o primeiro sitio da África do Sul a estar inscrito na Lista de Patrimonios Mundiais, e uma das joias da zona costeira do pais, com um mosaico absolutamente unico de ecossistemas – pantanos, lagos e lagoas, praias, recifes de corais, terras pantanosas, florestas costeiras e savana – enriquecidos por uma enorme diversidade de vida animal, aves e especies marinhas.Reporter infoSA, incorporando material de Turismo Sul Africano
How OKR’s Completely Transformed Our Culture Will Development Eventually Make Itself Obsolete? End-of-Life Software: Keep it, Update it, or Fi… Guest author Scott Gerber is the founder of the Young Entrepreneur Council.As a company grows, individual teams often develop their own project management styles. But when you’re trying to scale, having multiple systems can lead to confusion and missed deadlines. So how do you adjust that methodology as you expand? See also: 11 Ways To Manage A Remote Team I asked a group of entrepreneurs from YEC who’ve been through the process to tell us more about how they handled this situation. Below, they offer their best tips for improving communication and transitioning seamlessly. Find The One Right ToolUsing multiple project management solutions across different internal groups is a recipe for disaster. You’ll find your teams acclimating to their individual workflows, and inevitably, your ability to pivot and push your entire company towards one platform will become more and more problematic. Dig in, do the research, find the right tool and commit to it. Although there are going to be obstacles to overcome, once your team has learned how to properly utilize your unified platform, you’ll have a more efficient, more communicative group of individuals. There are some great applications available for bringing complicated workflows together. We use Atlassian products like Jira, Confluence and BitBucket, due to their ability for link deployment, project management and business development. —Blair Thomas, EMerchantBrokerIdentify The Best And The Second Best Methodology For Your TeamsSince you’re growing, you should not use multiple systems and methodology. You have to pick the best project management tool and methodology that works for you and your clients/product. However, one tool and methodology doesn’t always work for all situations, so you want to identify a second best tool and methodology as well. For example, we use Jira to build products in Agile Methodology. It allows everyone to continuously update features, feedback, timeline and cost.For waterfall methodology, we use Basecamp, where features are defined in milestones with estimated time and cost. In both cases, processes are very clear to employees and the product owner. Our client and staff always come up with new tools and methodology; we listen to them to learn more, but still stick with these two for ourselves. —Piyush Jain, SIMpalmGet On A Single PlatformWhatever project management platform you choose, pick one.First, you’ll appreciate how training new hires is much more efficient because you actually have fewer systems to train them on, which also means fewer processes to learn, understand and eventually master. Next, you’ll realize how your project information and time-tracking data will become more accurate because there is no need for data re-entry. Finally, with all of your projects in one place, you’ll get more accurate project data and gain more insight into operations through reporting and visual dashboards. —David Ciccarelli, Voices.comAdopt A Uniform LanguageThe best team will have communication across the board, not within inner cliques of departments. At LexION Capital, I’ve used a uniform task management program company wide, and have seen great results. For incredibly busy employees, at best different communication styles are a time sink, and at worst they will create sealed lips. I adopted an easy “language” that the whole team can speak, and you should do the same.To take it a step further, I encourage cross-department work—you’ll see amazing results when projects get a fresh perspective. —Elle Kaplan, LexION CapitalStay Consistent Across SystemsI think that multiple project management systems are hard to manage as you scale. We have made that mistake before and it hinders productivity. I suggest having one main system for each group if they require different information.For example, we use GitHub for our technology team and Salesforce for our Sales Reps. It’s important to have all relevant information pertaining to each team located on the same platform. —Jayna Cooke, EVENTupDefine A Core Process And AdjustWhen the department is small, it’s okay to have variance across teams. As you scale, a more unified process enables efficiency and consistency gains. Rather than define a rigid process, work with your team to develop a project management core that can be expanded or trimmed to fit specific project needs. Then, select tooling that best supports the core. —Bryan Delaney, Skookum Digital WorksKeep The Platform Appropriate To The DivisionAt Hubstaff, we use two separate project management systems and one platform for communication. We find that Trello is more appropriate for marketing since we can see everything and collaborate openly, but PivotalTracker is better for our dev teams, since it’s suited to tracking technical issues. Each division has different things to work on, and that means specialized project management systems are ideal. Customer support and sales may need something with a robust CRM system, while HR will need an extremely secure platform. It all depends on the department. The one place where we insist on company-wide unity is our communications platform, which is Skype. We use this sparingly, but we do expect everyone to be online and available while they’re working. —Dave Nevogt, Hubstaff.comHave A Dashboard That Covers The Whole BusinessI think it’s okay for teams to utilize their favorite or most efficient project management systems. However, any business needs to have at least one dashboard where they can view progress and updates from all of the teams, all in one place. You don’t need all of the nitty gritty details, but the business owner should be able to get a high-level update on progress happening within all teams in the company.Your project manager or director should work with you to determine the goals or objectives to be tracked on the central business dashboard. It’s this person’s job to keep this dashboard updated so the business owner is aware of progress or potential issues. We use Google spreadsheets to track progress and updates for each department all in one place. —Andy Karuza, brandbuddeeEvery Tool Should Have A PurposeAt any company one can have multiple tools. But every tool should have a purpose — try to not have too many tools with overlapping functionalities. At Kvantum, we use Smartsheet for high-level project planning for product development, while the technical sprints are planned in YouTrack and the sales team uses Nimble. We also use Trello to manage day-to-day tasks, and built excel-based dashboards for leadership when it comes to HR, Finance, etc.It’s important to provide the purpose of using a tool and a proper workflow related to the tool to ensure its adoption across all teams. At Kvantum, teams are not required to use all of the tools we roll out. They just use what makes sense, and there is rarely any overlap between the tools we use in terms of purpose or workflow. —Shilpi Sharma, Kvantum Inc. Lead photo from Bigstock Photo, courtesy of Stephen Moyers Related Posts Tips for Selling Smart Supply Chain Solutions scott gerber Tags:#business#Guest Posts#Management#Project Management
Scaling applications to hundreds or thousands of servers is a common practice in today’s High-Performance Computing (HPC) centers. Intel® Omni-Path Architecture (Intel® OPA) is a leading interconnect used to connect servers (commonly referred to as nodes) in a data center to one another. The data throughput capability of Intel OPA in its first generation is 100 Gigabits per second in each direction of the link. This corresponds to up to 12.5 Gigabytes per second of uni-directional bandwidth and 25 Gigabytes of bi-directional bandwidth. In addition to high bandwidth, Intel OPA is a low latency and highly resilient interconnect with many different Quality of Service (QoS) features. One of these features is adaptive routing. In this article, we show exactly how to enable and test adaptive routing with a simple micro benchmark, and in a future article, we show its impact on application performance in a cluster environment. Stay tuned also for articles on other QoS features such as Traffic Flow Optimization, where the fabric intelligently prioritizes latency-sensitive messages during simultaneous transmission of larger messages, or bulk data such as storage.When Intel OPA fabric is installed in a cluster, the nodes are often connected in a topology known as a “fat tree”. In this configuration, compute nodes are connected to “edge” or “leaf” switches, and these edge switches are in turn connected to “core” or “spine” switches. This is a two-tier fabric. With the 48 port radix design in the Intel® Omni-Path Edge Switch (100 series), a maximum of 1152 nodes can be connected in a cluster with full bisectional bandwidth. In this configuration, 48 edge switches are used in the first edge switch tier, connected to 24 switches in the second core tier. 24 hosts are connected to each of the 48 edges, and the remaining 24 ports on each edge are connected to each of the core switches, one inter-switch link (ISL) between each edge and core. In theory, this configuration is non-blocking and is capable of providing full bandwidth from any node in the cluster to any other unique node at a given time. However, static fat tree routing algorithms (which determine exactly which path is taken between any two node pairs) have limitations and full bandwidth is not always seen in practice.Static fat tree routing must provide routing paths for every node talking to every other node. Since a node talking to itself would typically use shared memory within the node, the number of potential routes R for a total node count of N is governed by the simple equation R=N(N-1)=N²-N. Since the number of routes increase as a function of N² but the number of ISLs only grow as a function of N, this means that there are scenarios where certain host communications will use the same ISL. Although customized routing rules can be implemented for a very specific communication pattern that would allow for full bandwidth throughout the cluster, the more general routing requirement needs to be in place for a general purpose HPC cluster. How can you tell exactly which routing path your communication will take through the fabric? Packaged tools such as opareport can be used in conjunction with the sending and destination nodes, for example: opareport -o route -S nodepat:”NodeA hfi1_0”:port:1 -D nodepat:”NodeB hfi1_0”:port:1Adaptive routing has the ability to detect when hosts are trying to use the same routes, and adjust the routes in real time to alleviate the congestion and potentially improve the bandwidth delivered to each node. Consider a simplified scenario with only two edge switches and two cores switches as illustrated in Figure 1. In this example we elect Node A to communicate to Node C, and Node B to communicate to Node D. This is representative of how a real HPC application might communicate. Based on the fat tree topology, all traffic needs to travel through either Core switch 1 or Core switch 2 to reach the destination node. To further simply the demonstration, we only connect each edge to each core switch with one ISL. Therefore, there are only two paths possible for any messages traveling between the node pairs. Monitoring the ISL traffic is easy with tools such as the Intel OPA Fabric Manager GUI, or tools such as opatop.Figure 1: Simplified Fat tree topologyThe performance test selected is Ohio State Microbenchmarks* version 5.3, osu_mbw_mr test (http://mvapich.cse.ohio-state.edu/benchmarks/Opens in a new window). This test returns the aggregate bandwidth and message rate between the number of nodes or CPU core pairs used in the test. In the source code, we modified the timing during the bandwidth calculation to use the maximum pair communication time instead of the average time. This prevents the code from returning bandwidth which is over maximum theoretical line rate of 12.5 GB/sec.We have chosen to use a Message Passing Interface (MPI) library as packaged with Intel OPA software release 10.3.1.0.22, which is Open MPI 1.10.4. However, any MPI library could be used. It is especially important to use the MPI packaged with the -hfi suffix to ensure Performance Scaled Messaging (PSM2) is used, which is the most performant code path for Intel OPA. If you compile your Open MPI yourself, make sure you include PSM in your configure command. Running this benchmark between Node A and Node C, using one MPI rank on each node, returns the following result:We have omitted the first lines of output for the smaller message sizes for brevity. The bandwidth achieved between two nodes using one core per node is 12.38GB/s (uni-directional), which is 99% of the theoretical line rate of Intel OPA in its first generation. We expect when running between two node pairs (again, one core per node), the reported bandwidth should be close to double the bandwidth measured for one node pair. However, for static routing, that is not always the case as we will now show. In this test, in addition to Node A sending to Node C, now Node B is sending to Node D and the aggregate throughput is reported:Note that the order of the node list is important, as the first half of the nodes are identified as the “sending” nodes, or group 1, and the second half of the nodes are the corresponding “receiving” nodes in group 2. As seen by the output, it is only slightly higher than the single node pair. For this demonstration, we purposely selected these nodes because opareport told us that they share the same static route through Core switch 2 (see the red path in Figure 1). During the above four node benchmark test, Figure 2 is the output of “opatop” (0 → W → D keystrokes), showing 100% ISL utilization through Core switch 2 and only 50% utilization for the host ports, because two host pairs are sharing the single ISL routed through Core switch 2. Note that core switch 1 is not utilized, so it is not listed since opatop automatically sorts by the highest utilization, or may other available metrics.Figure 2: Output of opatop during 4 node test with static routing. Only the red path in Figure 1 is used.Let us repeat the above test, this time with adaptive routing enabled. To enable adaptive routing, simply toggle
What do Clint Eastwood and the Leaning Tower of Pisa have in common? Not much, as far as your brain is concerned. But take a look at a picture of the famous actor and landmark together, and your brain will link the two, thanks to neurons that rapidly encode associations between people and places, according to a new study. The discovery is a “first step” toward understanding how the brain encodes complex, movielike memories of past events, says study author Itzhak Fried, a neuroscientist at the University of California, Los Angeles. About 10 years ago, Fried and colleagues discovered a bizarre phenomenon. While probing a deep brain region called the medial temporal lobe (MTL) in people with epilepsy, who had part of their skulls temporarily removed so physicians could pinpoint the source of their seizures, they discovered a single neuron that started to fire like crazy whenever the patient saw a photograph of actress Jennifer Aniston. The team went on to show that other individual neurons in the same region—which includes the hippocampus, a structure long known to be vital to memory processing—responded to different celebrities, such as Julia Roberts and Halle Berry, and even specific events in 5-second clips from The Simpsons.The now-famous “Jennifer Aniston” neuron supports a widely held hypothesis that specific brain cells in this region encode discrete representations of places, people, and objects. These neurons have an “amazing” property, called invariance, which sensitizes them not just to one image of their “preferred” stimulus, but to many different versions, Fried explains. In his 2005 experiments, for example, the Aniston neuron responded to images of the actress in any outfit and with any haircut. The only image the cell did not respond to, oddly enough, was of the actress holding hands with Brad Pitt. It’s not that just one cell responds to Aniston’s image, Fried emphasizes. Thousands, if not millions, of other cells in the brain may also be sensitive to the actress. But the cells are so sparsely distributed in the region that researchers only pick up one or several at a time, he says.Sign up for our daily newsletterGet more great content like this delivered right to you!Country *AfghanistanAland IslandsAlbaniaAlgeriaAndorraAngolaAnguillaAntarcticaAntigua and BarbudaArgentinaArmeniaArubaAustraliaAustriaAzerbaijanBahamasBahrainBangladeshBarbadosBelarusBelgiumBelizeBeninBermudaBhutanBolivia, Plurinational State ofBonaire, Sint Eustatius and SabaBosnia and HerzegovinaBotswanaBouvet IslandBrazilBritish Indian Ocean TerritoryBrunei DarussalamBulgariaBurkina FasoBurundiCambodiaCameroonCanadaCape VerdeCayman IslandsCentral African RepublicChadChileChinaChristmas IslandCocos (Keeling) IslandsColombiaComorosCongoCongo, The Democratic Republic of theCook IslandsCosta RicaCote D’IvoireCroatiaCubaCuraçaoCyprusCzech RepublicDenmarkDjiboutiDominicaDominican RepublicEcuadorEgyptEl SalvadorEquatorial GuineaEritreaEstoniaEthiopiaFalkland Islands (Malvinas)Faroe IslandsFijiFinlandFranceFrench GuianaFrench PolynesiaFrench Southern TerritoriesGabonGambiaGeorgiaGermanyGhanaGibraltarGreeceGreenlandGrenadaGuadeloupeGuatemalaGuernseyGuineaGuinea-BissauGuyanaHaitiHeard Island and Mcdonald IslandsHoly See (Vatican City State)HondurasHong KongHungaryIcelandIndiaIndonesiaIran, Islamic Republic ofIraqIrelandIsle of ManIsraelItalyJamaicaJapanJerseyJordanKazakhstanKenyaKiribatiKorea, Democratic People’s Republic ofKorea, Republic ofKuwaitKyrgyzstanLao People’s Democratic RepublicLatviaLebanonLesothoLiberiaLibyan Arab JamahiriyaLiechtensteinLithuaniaLuxembourgMacaoMacedonia, The Former Yugoslav Republic ofMadagascarMalawiMalaysiaMaldivesMaliMaltaMartiniqueMauritaniaMauritiusMayotteMexicoMoldova, Republic ofMonacoMongoliaMontenegroMontserratMoroccoMozambiqueMyanmarNamibiaNauruNepalNetherlandsNew CaledoniaNew ZealandNicaraguaNigerNigeriaNiueNorfolk IslandNorwayOmanPakistanPalestinianPanamaPapua New GuineaParaguayPeruPhilippinesPitcairnPolandPortugalQatarReunionRomaniaRussian FederationRWANDASaint Barthélemy Saint Helena, Ascension and Tristan da CunhaSaint Kitts and NevisSaint LuciaSaint Martin (French part)Saint Pierre and MiquelonSaint Vincent and the GrenadinesSamoaSan MarinoSao Tome and PrincipeSaudi ArabiaSenegalSerbiaSeychellesSierra LeoneSingaporeSint Maarten (Dutch part)SlovakiaSloveniaSolomon IslandsSomaliaSouth AfricaSouth Georgia and the South Sandwich IslandsSouth SudanSpainSri LankaSudanSurinameSvalbard and Jan MayenSwazilandSwedenSwitzerlandSyrian Arab RepublicTaiwanTajikistanTanzania, United Republic ofThailandTimor-LesteTogoTokelauTongaTrinidad and TobagoTunisiaTurkeyTurkmenistanTurks and Caicos IslandsTuvaluUgandaUkraineUnited Arab EmiratesUnited KingdomUnited StatesUruguayUzbekistanVanuatuVenezuela, Bolivarian Republic ofVietnamVirgin Islands, BritishWallis and FutunaWestern SaharaYemenZambiaZimbabweI also wish to receive emails from AAAS/Science and Science advertisers, including information on products, services and special offers which may include but are not limited to news, careers information & upcoming events.Required fields are included by an asterisk(*)How such neurons contribute to the seamless, movielike recollections of past events—called episodic memories—is still largely unknown, however. So in the new study, Fried and colleagues showed 14 people undergoing exploratory surgery for epilepsy 100 to 200 randomly ordered images, including pictures of their loved ones, celebrities such as Clint Eastwood, and volleyball player Kerri Walsh, as well as landmarks such as the Eiffel Tower, the White House, and the Hollywood sign in Los Angeles. Each patient already had electrodes sunk deep into their MTL to detect aberrant electrical activity, and scientists used the wires to listen for cells that fired up in response to different images.Of the roughly 600 neurons the team recorded in each patient, between 2 and 28 cells fired vigorously in response to at least one image. Next, the researchers presented participants with digitally altered photographs in which a neuron’s “preferred” image, such as a photo of Clint Eastwood, was superimposed on a background the neuron had ignored in a previous trial, such as the Leaning Tower of Pisa. In a series of memory tasks, the participants were asked to match pairs of separate images based on the doctored composites. If they’d seen the Eastwood composite, for example, their task might be to pair a photo of Eastwood with a separate photo of the tower.Even after one exposure to the composites, neurons that had previously fired exclusively in response to one picture—like that of Eastwood—significantly increased their firing rate when exposed to the image with which it had been combined—in one case, by 230%, Fried and colleagues report today in the journal Neuron. The fact that an individual neuron can adapt its firing rate so quickly could help explain how large, dynamic neuronal networks form complicated memories of past events, Fried says.The findings are “very consistent with results from a number of animal studies” that show rapid changes in hippocampal neural activity during learning, says Loren Frank, a neuroscientist at the University of California, San Francisco, who was not involved in the study. “This is indeed the sort of thing that has to happen” in the brain to store memories of once-in-a-lifetime events, he says.Fried hopes the new study will contribute to efforts to restore memory in people suffering from traumatic brain injury or Alzheimer’s disease, such as an ambitious Defense Administration Research Project Agency (DARPA)-funded initiative aimed at restoring memory function in neurological patients in which his lab is participating. Many scientists have expressed skepticism about the DARPA project, and Fried notes that bridging the gap between basic research on simple associations and clinical memory treatments is “a formidable challenge.” Such projects, he says, “should avoid the pitfalls of unrealistic expectations for early ‘big wins.’ ”
The content of this field is kept private and will not be shown publicly.
Royal Challengers Bangalore (RCB) pacer Adam Milne, who is out with an injury, was on Wednesday replaced by Sreenath Arvind.The technical committee of the Indian Premier League (IPL) approved the change on Wednesday.This will be Arvind’s second stint with RCB. The left-handed paceman was part of the RCB squad in the 2011 and 2012 seasons and was their highest wicket-taker in 2011 with 21 scalps in 13 games.The technical committee consists of Board of Control for Cricket in India (BCCI) secretary Anurag Thakur, Sourav Dasgupta, Subir Ganguly, Sourav Ganguly and Ravi Shastri.