GPM合并资料整理-GEM部分-Toy模板网

这篇具有很好参考价值的文章主要介绍了GPM合并资料整理-GEM部分。希望对大家有所帮助。如果存在错误或未考虑完全的地方，请大家不吝赐教，您也可以点击"举报违法"按钮提交疑问。

一、性能数据上报项

1. CPU模块

上报键值	说明	采集平台
cpu	当前进程cpu使用率平均值	Android & iOS
totcpu	系统cpu总使用率平均值	Android & iOS
cpu_temp_max	cpu最高温度	Android
cpu_temp_avg	cpu温度平均值	Android
gpu_temp_avg	gpu温度平均值	Android
gpu_temp_max	gpu最高温度	Android
gpu	gpu使用率平均值	Android
gpu_model	gpu型号	Android
cpu_model	cpu型号	Android
cpu_core	cpu核心数	Android
cpu_freq	cpu最大时钟频率	Android

2. memory及存储模块

上报键值	说明	采集平台
mem	当前进程内存使用量平均值	Android & iOS
availmem	当前设备可用内存量平均值	Android & iOS
max_mem	内存使用量最大值	Android & iOS
start_mem	开始对局时内存使用量	Android & iOS
end_mem	结束对局时内存使用量	Android & iOS
totalmem	当前设备总内存量	Android & iOS
total_storage	当前设备总存储空间大小	Android & iOS
free_storage	当前设备可用存储空间大小	Android & iOS

3. 网卡模块

上报键值	说明	采集平台
wsndpkt	无线网卡发包总数	Android & iOS
wrcvpkt	无线网卡收包总数	Android & iOS
wsnddrop	无线网卡发包丢包	Android & iOS
wrcvdrop	无线网卡收包丢包	Android & iOS
wsnderr	无线网卡发包错包	Android & iOS
wrcverr	无线网卡收包错包	Android & iOS
msndpkt	移动网卡发包总数	Android & iOS
mrcvpkt	移动网卡收包总数	Android & iOS
msnddrop	移动网卡发包丢包	Android & iOS
mrcvdrop	移动网卡收包丢包	Android & iOS
msnderr	移动网卡发包错包	Android & iOS
mrcverr	移动网卡收包错包	Android & iOS

4. FPS模块

上报键值	说明	采集平台
fmin	fps最小值	Android & iOS
fmax	fps最大值	Android & iOS
favg	fps平均值	Android & iOS
fheavy	fps严重抖动个数（本次fps值比上次fps值低10以上）	Android & iOS
flight	fps轻微抖动个数（本次fps值比上次fps值低4以上，10以下）	Android & iOS
ftotal	fps采集总个数	Android & iOS
fcntx0	fps自定义抖动个数（阈值由云控下发）	Android & iOS
lfps1	低于云控下发的lfps1阈值的个数	Android & iOS
lfps2	低于云控下发的lfps2阈值的个数	Android & iOS
lfps3	低于云控下发的lfps3阈值的个数	Android & iOS

5. 电量模块

上报键值	说明	采集平台
battery	电量消耗（开始-结束）	Android & iOS
start_battery	开始对局时电量值	Android & iOS
end_battery	结束对局时电量值	Android & iOS
bs	对局过程中是否有充电操作	Android & iOS
bt	电池温度平均值	Android
max_battery_temp	电池温度最大值	Android

6. 网络相关模块

上报键值	说明	采集平台
netflow	流量消耗	Android & iOS
devices	当前wifi下连接设备数	Android & iOS
wifi_num	wifi个数	Android
wifi_rssi	当前wifi信号强度	Android
wifi_speed	当前wifi链接速度	Android
gate_delay	到网关延迟（ping）	Android & iOS
signal_level	移动信号强度	Android
xg	当前网络类型	Android & iOS
ldns	本机DNS服务器	Android & iOS

7. 机型相关

上报键值	说明	采集平台
manufacturer	厂商	Android
brand	品牌	Android
model	型号	Android
resolution	分辨率	Android

cpu&memory采集具体实现代码

1. cpu

Android采集代码：

  public static float totalCpuUsageRate() {
  	return getTotalCpuUsageRate(USAGE_RATE_CAL_INTERVAL_MILLS_DEFAULT);
  }

  public static float getTotalCpuUsageRate(long calIntervalMills) {
  	if (0 > calIntervalMills) {
  		return VALUE_GET_FAILED;
  	}
  	float totalCpuUsageRate = VALUE_GET_FAILED;
  	try {
  		TotalCpuStatInfo totalCpuStatInfo1 = TotalCpuStatInfo.get();
  		long cpuTotal1 = totalCpuStatInfo1.getCpuTotal();
  		if (0 > cpuTotal1) {
  			return totalCpuUsageRate;
  		}
  		long total1 = totalCpuStatInfo1.getTotal();
  		if (0 > total1) {
  			return totalCpuUsageRate;
  		}
  		try {
  			Thread.sleep(calIntervalMills);
  		} catch (Exception e) {
  			GPMLogger.w(e, "getTotalCpuUsageRate failed");
  		}
  		TotalCpuStatInfo totalCpuStatInfo2 = TotalCpuStatInfo.get();
  		long cpuTotal2 = totalCpuStatInfo2.getCpuTotal();
  		if (0 > cpuTotal2) {
  			return totalCpuUsageRate;
  		}
  		long total2 = totalCpuStatInfo2.getTotal();
  		if (0 > total2) {
  			return totalCpuUsageRate;
  		}
  		long totalDiff = total2 - total1;
  		if (0 == totalDiff) {
  			// NOTE: totalDiff为分母
  			return totalCpuUsageRate;
  		}
  		long cpuTotalDiff = cpuTotal2 - cpuTotal1;
  		return 100 * cpuTotalDiff / (totalDiff + 0f);
  	} catch (Exception e) {
  		GPMLogger.w(e, "getTotalCpuUsageRate failed");
  	}
  		return totalCpuUsageRate;
  	}

  public static float myCpuUsageRate() {
  	return getMyCpuUsageRate(USAGE_RATE_CAL_INTERVAL_MILLS_DEFAULT);
  }

  public static float getMyCpuUsageRate(long calIntervalMills) {
  	if (0 > calIntervalMills) {
  		return VALUE_GET_FAILED;
  	}
  	float myCpuUsageRate = VALUE_GET_FAILED;
  	try {
  		TotalCpuStatInfo totalCpuStatInfo1 = TotalCpuStatInfo.get();
  		long total1 = totalCpuStatInfo1.getTotal();
  		if (0 > total1) {
  			return myCpuUsageRate;
  		}
  		MyCpuStatInfo myCpuStatInfo1 = MyCpuStatInfo.get();
  		long myCpuTotal1 = myCpuStatInfo1.getCpuTotal();
  		if (0 > myCpuTotal1) {
  			return myCpuUsageRate;
  		}
  		try {
  			Thread.sleep(calIntervalMills);
  		} catch (Exception e) {
  			GPMLogger.w(e, "getMyCpuUsageRate failed");
  		}
  		TotalCpuStatInfo totalCpuStatInfo2 = TotalCpuStatInfo.get();
  		long total2 = totalCpuStatInfo2.getTotal();
  		if (0 > total2) {
  			return myCpuUsageRate;
  		}
  		MyCpuStatInfo myCpuStatInfo2 = MyCpuStatInfo.get();
  		long myCpuTotal2 = myCpuStatInfo2.getCpuTotal();
  		if (0 > myCpuTotal2) {
  			return myCpuUsageRate;
  		}
  		long totalDiff = total2 - total1;
  		if (0 == totalDiff) {
  			// NOTE: totalDiff为分母
  			return myCpuUsageRate;
  		}
  		long myCpuTotalDiff = myCpuTotal2 - myCpuTotal1;
  		return 100 * myCpuTotalDiff / (totalDiff + 0f);
  	} catch (Exception e) {
  		GPMLogger.w(e, "getTotalCpuUsageRate failed");
  	}
  	return myCpuUsageRate;
  }

  private static class TotalCpuStatInfo {
  	private static final int MIN_CPU_STAT_ITEM_NUM = 9;
  	private static final String[] INVALID_RAW_CPU_STAT_ITEMS =
          new String[MIN_CPU_STAT_ITEM_NUM];

  	public final long user;
  	public final long nice;
  	public final long system;
  	public final long idle;
  	public final long iowait;
  	public final long irq;
  	public final long softirq;

  	public static TotalCpuStatInfo get() {
  		BufferedReader cpuStatInfoReader = null;
  		try {
  			cpuStatInfoReader = new BufferedReader(new InputStreamReader(new FileInputStream(STAT_ABS_DIR), UTF8_CHARSET));
  			return new TotalCpuStatInfo(cpuStatInfoReader.readLine());
  		} catch (Exception e) {
  			// NOTE: 高版本上获取不到, 不打印堆栈
  			GPMLogger.d("get total cpu stat info failed");
  			return new TotalCpuStatInfo(null);
  		} finally {
  			if (null != cpuStatInfoReader) {
  				try {
  					cpuStatInfoReader.close();
  				} catch (IOException ignored) {
  				}
  			}
  		}
  	}

  	private TotalCpuStatInfo(String cpuStatLine) {
  		String[] rawCpuStatItems = CpuStatInfoHelper.buildRawCpuStatItems(cpuStatLine, MIN_CPU_STAT_ITEM_NUM, INVALID_RAW_CPU_STAT_ITEMS);
  		user = CpuStatInfoHelper.buildCpuStatItem(rawCpuStatItems[2]);
  		nice = CpuStatInfoHelper.buildCpuStatItem(rawCpuStatItems[3]);
  		system = CpuStatInfoHelper.buildCpuStatItem(rawCpuStatItems[4]);
  		idle = CpuStatInfoHelper.buildCpuStatItem(rawCpuStatItems[5]);
  		iowait = CpuStatInfoHelper.buildCpuStatItem(rawCpuStatItems[6]);
  		irq = CpuStatInfoHelper.buildCpuStatItem(rawCpuStatItems[7]);
  		softirq = CpuStatInfoHelper.buildCpuStatItem(rawCpuStatItems[8]);
  	}

  	public long getCpuTotal() {
  		return user + nice + system + iowait + irq + softirq;
  	}

  	public long getTotal() {
  		return user + nice + system + idle + iowait + irq + softirq;
  	}
  }

  private static class MyCpuStatInfo {
  	private static final int MIN_CPU_STAT_ITEM_NUM = 17;
  	private static final String[] INVALID_RAW_CPU_STAT_ITEMS = new String[MIN_CPU_STAT_ITEM_NUM];

  	public final long utime;
  	public final long stime;
  	public final long cutime;
  	public final long cstime;
  	
  	public static MyCpuStatInfo get() {
  		BufferedReader cpuStatInfoReader = null;
  		try {
  			cpuStatInfoReader = new BufferedReader(new InputStreamReader(new FileInputStream(MY_STAT_ABS_DIR), UTF8_CHARSET));
  			return new MyCpuStatInfo(cpuStatInfoReader.readLine());
  		} catch (Exception e) {
  			GPMLogger.w(e, "get my cpu stat info failed");
  			return new MyCpuStatInfo(null);
  		} finally {
  			if (null != cpuStatInfoReader) {
  				try {
  					cpuStatInfoReader.close();
  				} catch (IOException ignored) {
  				}
  			}
  		}
  	}

  	private MyCpuStatInfo(String cpuStatLine) {
  		String[] rawCpuStatItems = CpuStatInfoHelper.buildRawCpuStatItems(cpuStatLine, MIN_CPU_STAT_ITEM_NUM, INVALID_RAW_CPU_STAT_ITEMS);
  		utime = CpuStatInfoHelper.buildCpuStatItem(rawCpuStatItems[13]);
  		stime = CpuStatInfoHelper.buildCpuStatItem(rawCpuStatItems[14]);
  		cutime = CpuStatInfoHelper.buildCpuStatItem(rawCpuStatItems[15]);
  		cstime = CpuStatInfoHelper.buildCpuStatItem(rawCpuStatItems[16]);
  	}

  	public long getCurrentThreadCpuTotal() {
  		return utime + stime;
  	}

  	public long getCpuTotal() {
  		return utime + stime + cutime + cstime;
  	}
  }

IOS采集代码：

  // ============当前进程cpu使用率============
  - (double) getCPU {
  	kern_return_t kr;
  	task_info_data_t tinfo;
  	mach_msg_type_number_t task_info_count;
  	task_info_count = TASK_INFO_MAX;
  	kr = task_info(mach_task_self(), TASK_BASIC_INFO, (task_info_t)tinfo, &task_info_count);
  	if (kr != KERN_SUCCESS) {
  		GSDKLOG(@"================CPU:%.2f",-1.0f);
  		return -1;
  	}
  	task_basic_info_t      basic_info;
  	thread_array_t         thread_list;
  	mach_msg_type_number_t thread_count;
  	thread_info_data_t     thinfo;
  	mach_msg_type_number_t thread_info_count;
  	thread_basic_info_t basic_info_th;
  	uint32_t stat_thread = 0; // Mach threads
  	basic_info = (task_basic_info_t)tinfo;
  	// get threads in the task
  	kr = task_threads(mach_task_self(), &thread_list, &thread_count);
  	if (kr != KERN_SUCCESS) {
  		GSDKLOG(@"=============CPU:%.2f",-1.0f);
  		return -1;
  	}
  	if (thread_count > 0)
  		stat_thread += thread_count;
  	long tot_sec = 0;
  	long tot_usec = 0;
  	float tot_cpu = 0;
  	int j;
  	for (j = 0; j < thread_count; j++) {
  		thread_info_count = THREAD_INFO_MAX;
  		kr = thread_info(thread_list[j], THREAD_BASIC_INFO, (thread_info_t)thinfo, &thread_info_count);
  		if (kr != KERN_SUCCESS) {
  			GSDKLOG(@"=========CPU:%.2f",-1.0f);
  			return -1;
  		}
  		basic_info_th = (thread_basic_info_t)thinfo;
  		if (!(basic_info_th->flags & TH_FLAGS_IDLE)) {
  			tot_sec = tot_sec + basic_info_th->user_time.seconds + basic_info_th->system_time.seconds;
  			tot_usec = tot_usec + basic_info_th->system_time.microseconds + basic_info_th->system_time.microseconds;
  			tot_cpu = tot_cpu + basic_info_th->cpu_usage / (float)TH_USAGE_SCALE * 100.0;
  		}
  	} // for each thread
  	kr = vm_deallocate(mach_task_self(), (vm_offset_t)thread_list, thread_count * sizeof(thread_t));
  	GSDKLOG(@"=============CPU:%.2f",tot_cpu);
  	return tot_cpu;
  }


  // ============当前进程cpu使用率============
  processor_info_array_t cpuInfo, prevCpuInfo;
  mach_msg_type_number_t numCpuInfo, numPrevCpuInfo;
  unsigned numCPUs;
  NSLock * CPUUsageLock;

  - (void) getSystemCPU {
  	int mib[2U] = { CTL_HW, HW_NCPU };
  	size_t sizeOfNumCPUs = sizeof(numCPUs);
  	int status = sysctl(mib, 2U, &numCPUs, &sizeOfNumCPUs, NULL, 0U);
  	if (status)
  		numCPUs = 1;
  	CPUUsageLock = [[NSLock alloc] init];
  }

  - (double) getSystemCPUCircle {
  	natural_t numCPUsU = 0U;
  	kern_return_t err = host_processor_info(mach_host_self(), PROCESSOR_CPU_LOAD_INFO, &numCPUsU, &cpuInfo, &numCpuInfo);
  	if (err == KERN_SUCCESS) {
  		[CPUUsageLock lock];
  		float tot_inUse = 0;
  		float tot_total = 0;
  		for(unsigned i = 0U; i < numCPUs; ++i) {
  			float inUse, total;
  			if(prevCpuInfo) {
  				inUse = ((cpuInfo[(CPU_STATE_MAX * i) + CPU_STATE_USER] - prevCpuInfo[(CPU_STATE_MAX * i) + CPU_STATE_USER])
                   + (cpuInfo[(CPU_STATE_MAX * i) + CPU_STATE_SYSTEM] - prevCpuInfo[(CPU_STATE_MAX * i) + CPU_STATE_SYSTEM])
                   + (cpuInfo[(CPU_STATE_MAX * i) + CPU_STATE_NICE]   - prevCpuInfo[(CPU_STATE_MAX * i) + CPU_STATE_NICE]));
  				total = inUse + (cpuInfo[(CPU_STATE_MAX * i) + CPU_STATE_IDLE] - prevCpuInfo[(CPU_STATE_MAX * i) + CPU_STATE_IDLE]);
  			} else {
  				inUse = cpuInfo[(CPU_STATE_MAX * i) + CPU_STATE_USER] + cpuInfo[(CPU_STATE_MAX * i) + CPU_STATE_SYSTEM] + cpuInfo[(CPU_STATE_MAX * i) + CPU_STATE_NICE];
  				total = inUse + cpuInfo[(CPU_STATE_MAX * i) + CPU_STATE_IDLE];
  			}
  			tot_inUse += inUse;
  			tot_total += total;
  		}
  		[CPUUsageLock unlock];
  		if (prevCpuInfo) {
  			size_t prevCpuInfoSize = sizeof(integer_t) * numPrevCpuInfo;
  			vm_deallocate(mach_task_self(), (vm_address_t)prevCpuInfo, prevCpuInfoSize);
  		}
  
  		prevCpuInfo = cpuInfo;
  		numPrevCpuInfo = numCpuInfo;
  
  		cpuInfo = NULL;
  		numCpuInfo = 0U;
  		if (tot_total > 0) {
  			GSDKLOG(@"CPU Usage: %.2f%%",tot_inUse/tot_total * 100.f);
  			return tot_inUse/tot_total * 100.f;
  		} else {
  			return -1;
  		}
  	} else {
  		GSDKLOG(@"Error!");
  		return -1;
  	}
  }

2. memory

Android采集代码：

  public static long getAvailableMemory(Context context) {
  	if (null == context) {
  		return VALUE_GET_FAILED;
  	}
  	@SuppressWarnings("AlibabaLowerCamelCaseVariableNaming")
  	long availableMemoryInMB = VALUE_GET_FAILED;
  	try {
  		ActivityManager activityManager =
              (ActivityManager) context.getSystemService(Context.ACTIVITY_SERVICE);
  		if (null == activityManager) {
  			return availableMemoryInMB;
  		}
  		ActivityManager.MemoryInfo memoryInfo = new ActivityManager.MemoryInfo();
  		activityManager.getMemoryInfo(memoryInfo);
  		availableMemoryInMB = memoryInfo.availMem / (1024 * 1024);
  	} catch (Exception e) {
  		GPMLogger.w(e, "availableMemoryInMB failed");
  	}
  	return availableMemoryInMB;
  }
  
  // ============当前设备可用内存大小（原方案）============
  public static long getProcessMemery() {
  	// TODO: 厂商反馈Debug.getPss()计算还是过于复杂，后续考虑使用Process.getPss()或者仿照Process.getPss()实现
  	// NOTE: 当前实现应该在低优先级后台线程调用
  	if (Build.VERSION_CODES.ICE_CREAM_SANDWICH <= Build.VERSION.SDK_INT) {
  		return Debug.getPss() / 1024;
  	}
  	// NOTE: 仅在Android14下使用，不涉及私有API访问限制
  	// Android 2.3上Process已有getPss方法
  	// https://android.googlesource.com/platform/frameworks/base/+/refs/heads/gingerbread/core/java/android/os/Process.java
  	try {
  		Method getPssMethod = Process.class.getMethod(GET_PSS_METHOD_NAME, int.class);
  		return ((long) getPssMethod.invoke(null, Process.myPid())) / 1024;
  	} catch (Exception e) {
  		return VALUE_GET_FAILED;
  	}
  }
  // ============当前设备可用内存大小（优化后方案）============
  private static ActivityManager sActivityMgr = null;
  private static int sPid = 0;

  public static int getProcessMemery(Context context) {
  	try {
  		if (sActivityMgr == null) {
  			sActivityMgr = (ActivityManager) context.getSystemService(context.ACTIVITY_SERVICE);
  		}

  		if (sPid == 0) {
  			sPid = android.os.Process.myPid();
  		}

  		if (sActivityMgr == null || sPid == 0) {
  			return -1;
  		}

  		int[] myMempid = new int[]{sPid};
  		Debug.MemoryInfo[] memoryInfo = sActivityMgr.getProcessMemoryInfo(myMempid);
  		int memSize = 0;
  		if (memoryInfo != null && memoryInfo.length > 0) {
  			memSize = memoryInfo[0].getTotalPss();
  		}
  		return memSize;
  	} catch (Exception e) {
  		Logger.e("PSS fetch error: " + e.getMessage());
  	}
  	return -1;
  }

IOS采集代码：

  // ============当前设备可用内存大小============
  - (double) getSystemAvailableMemory {
  	// 获取当前设备可用内存(单位：MB）
  	vm_statistics_data_t vmStats;
  	mach_msg_type_number_t infoCount = HOST_VM_INFO_COUNT;
  	kern_return_t kernReturn = host_statistics(mach_host_self(), HOST_VM_INFO, (host_info_t)&vmStats, &infoCount);
  	if (kernReturn != KERN_SUCCESS) {
  		GSDKLOG(@"AvailableMemory=====%.2f",-1.0f);
  		return -1;
  	}

  	double availMem = (vm_page_size * vmStats.free_count + vm_page_size * vmStats.inactive_count) / 1024.0 / 1024.0;
  	GSDKLOG(@"AvailableMemory=====%.2f",availMem);
  	return availMem;
  }

  // ============当前进程使用内存大小（原方案）============
  - (double) getFootPrint {
  	task_vm_info_data_t vmInfo;
  	mach_msg_type_number_t count = TASK_VM_INFO_COUNT;
  	kern_return_t result = task_info(mach_task_self(), TASK_VM_INFO, (task_info_t) &vmInfo, &count);
  	if (result != KERN_SUCCESS) {
  		GSDKLOG(@"FootPrint=====%.2f",-1.0f);
  		return -1;
  	}
  	GSDKLOG(@"FootPrint=====%i",(int)((vmInfo.phys_footprint) / 1024.0 / 1024.0));
  	return vmInfo.phys_footprint / 1024.0 / 1024.0;
  }

  // ============当前进程使用内存大小（优化后方案）============
  - (instancetype) init {
  	if (self = [super init]) {
  		_isLowLevelDevice = NO;
  		struct utsname systemInfo;
  		uname(&systemInfo);
  		NSString *platform = [NSString stringWithCString:systemInfo.machine encoding:NSASCIIStringEncoding];
  		if ([platform isEqualToString:@"iPhone7,1"] || [platform isEqualToString:@"iPhone7,2"] || [platform isEqualToString:@"iPhone6,1"] || [platform isEqualToString:@"iPhone6,2"]) {
  			_isLowLevelDevice = YES;
  		}
  	}
  	return self;
  }
  
  - (double) getFootPrint {
  	task_vm_info_data_t vmInfo;
  	mach_msg_type_number_t count = TASK_VM_INFO_COUNT;
  	kern_return_t result = task_info(mach_task_self(), TASK_VM_INFO, (task_info_t) &vmInfo, &count);
  	if (result != KERN_SUCCESS) {
  		GSDKLOG(@"FootPrint=====%.2f",-1.0f);
  		return -1;
  	}
  	double footPrint = -1;
  	if (_isLowLevelDevice || [UIDevice.currentDevice.systemVersion doubleValue] >= 12.0) {
  		footPrint = vmInfo.phys_footprint / 1024.0 / 1024.0;
  	}
  	if ([UIDevice.currentDevice.systemVersion doubleValue] < 12.0) {
  		footPrint = (vmInfo.internal + vmInfo.compressed - vmInfo.purgeable_volatile_pmap) / 1024.0 / 1024.0;
  	}
  	GSDKLOG(@"FootPrint=====%i",int((vmInfo.phys_footprint) / 1024.0 / 1024.0));
  	return footPrint;
  }

二、性能数据统一上报类示例

使用同一个TDMReportHelper实例进行上报即可：文章来源地址https://www.toymoban.com/news/detail-781086.html

TDMReportHelper eventReportHandler = new TDMReportHelper(eventName);
eventReportHandler.addSS(key, value);
eventReportHandler.report();
eventReportHandler.destory();

到了这里，关于GPM合并资料整理-GEM部分的文章就介绍完了。如果您还想了解更多内容，请在右上角搜索TOY模板网以前的文章或继续浏览下面的相关文章，希望大家以后多多支持TOY模板网！