diff --git a/.gitmodules b/.gitmodules
index e6a012fbe..b60b989ef 100644
--- a/.gitmodules
+++ b/.gitmodules
@@ -7,3 +7,7 @@
 [submodule "tpl/RAJA"]
 	path = tpl/RAJA
 	url = https://github.com/LLNL/RAJA.git
+[submodule "tpl/RAJAvec"]
+	path = tpl/RAJAvec
+	url = https://github.com/LLNL/RAJA.git
+	branch = feature/kunen1/vector
diff --git a/CMakeLists.txt b/CMakeLists.txt
index 75d53ca9a..1e907c8b6 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -13,14 +13,21 @@ cmake_minimum_required(VERSION 3.9)
 option(ENABLE_RAJA_SEQUENTIAL "Run sequential variants of RAJA kernels. Disable
 this, and all other variants, to run _only_ raw C loops." On)
 
+option(ENABLE_RAJA_VECTORIZATION "Run vectorized variants of RAJA kernels. Disable
+this, and all other variants, to run _only_ raw C loops." Off)
+
 #
 # Initialize the BLT build system
 #
 
+message(STATUS ${ENABLE_RAJA_SEQUENTIAL})
+message(STATUS ${ENABLE_RAJA_VECTORIZATION})
+
 if (PERFSUITE_ENABLE_WARNINGS)
   set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wall -Wextra -Werror")
 endif()
 
+set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mavx2")
 set(ENABLE_TESTS Off CACHE BOOL "Enable BLT and RAJA tests")
 set(ENABLE_EXAMPLES Off CACHE BOOL "Enable RAJA examples")
 set(ENABLE_EXERCISES Off CACHE BOOL "Enable RAJA exercises")
@@ -48,12 +55,6 @@ set(RAJA_DATA_ALIGN 64)
 
 
 # exclude RAJA make targets from top-level build...
-add_subdirectory(tpl/RAJA)
-
-get_property(RAJA_INCLUDE_DIRS DIRECTORY tpl/RAJA PROPERTY INCLUDE_DIRECTORIES)
-include_directories(${RAJA_INCLUDE_DIRS})
-
-
 #
 # Setup variables to pass to Perf suite
 #
@@ -63,14 +64,23 @@ include_directories(${RAJA_INCLUDE_DIRS})
 # performance issues in the xl compiler.
 #
 if (ENABLE_RAJA_SEQUENTIAL)
+  add_subdirectory(tpl/RAJAvec)
+  get_property(RAJA_INCLUDE_DIRS DIRECTORY tpl/RAJAvec PROPERTY INCLUDE_DIRECTORIES)
+  include_directories(${RAJA_INCLUDE_DIRS})
   add_definitions(-DRUN_RAJA_SEQ)
 endif ()
+#if(ENABLE_RAJA_VECTORIZATION)
+#  add_subdirectory(tpl/RAJAvec)
+#  get_property(RAJA_INCLUDE_DIRS DIRECTORY tpl/RAJAvec PROPERTY INCLUDE_DIRECTORIES)
+#  include_directories(${RAJA_INCLUDE_DIRS})
+#  add_definitions(-DRUN_RAJA_VEC)
+#endif ()
 if (ENABLE_OPENMP)
   add_definitions(-DRUN_OPENMP)
 endif ()
 
 set(RAJA_PERFSUITE_VERSION_MAJOR 0)
-set(RAJA_PERFSUITE_VERSION_MINOR 8)
+set(RAJA_PERFSUITE_VERSION_MINOR 9)
 set(RAJA_PERFSUITE_VERSION_PATCHLEVEL 0)
 
 set(RAJA_PERFSUITE_DEPENDS RAJA)
diff --git a/scripts/lc-builds/blueos_xl-2020.09.17.sh b/scripts/lc-builds/blueos_xl-2020.09.17.sh
index 1205c5cae..67bb5e441 100755
--- a/scripts/lc-builds/blueos_xl-2020.09.17.sh
+++ b/scripts/lc-builds/blueos_xl-2020.09.17.sh
@@ -10,16 +10,27 @@
 BUILD_SUFFIX=lc_blueos-xl_2020.09.17
 RAJA_HOSTCONFIG=../tpl/RAJA/host-configs/lc-builds/blueos/xl_X.cmake
 
-rm -rf build_${BUILD_SUFFIX} 2>/dev/null
-mkdir build_${BUILD_SUFFIX} && cd build_${BUILD_SUFFIX}
+rm -rf build_${BUILD_SUFFIX}_$1 2>/dev/null
+mkdir build_${BUILD_SUFFIX}_$1 && cd build_${BUILD_SUFFIX}_$1
 
 module load cmake/3.14.5
 
+if [ "$1" == "seq" ]; then
+    argS="On"
+    argV="Off"
+    RAJA_HOSTCONFIG=../tpl/RAJA/host-configs/lc-builds/toss3/clang_X.cmake
+else
+    argS="On"
+    argV="On"
+    RAJA_HOSTCONFIG=../tpl/RAJAvec/host-configs/lc-builds/toss3/clang_X.cmake
+fi
+
 cmake \
   -DCMAKE_BUILD_TYPE=Release \
   -DCMAKE_CXX_COMPILER=/usr/tce/packages/xl/xl-2020.09.17/bin/xlc++_r \
   -C ${RAJA_HOSTCONFIG} \
   -DENABLE_OPENMP=On \
+  -DENABLE_RAJA_VECTORIZATION=$argV \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss3_clang10.0.1.sh b/scripts/lc-builds/toss3_clang10.0.1.sh
index 9b069228e..cc0a5bc43 100755
--- a/scripts/lc-builds/toss3_clang10.0.1.sh
+++ b/scripts/lc-builds/toss3_clang10.0.1.sh
@@ -8,18 +8,30 @@
 ###############################################################################
 
 BUILD_SUFFIX=lc_toss3-clang-10.0.1
-RAJA_HOSTCONFIG=../tpl/RAJA/host-configs/lc-builds/toss3/clang_X.cmake
 
-rm -rf build_${BUILD_SUFFIX} 2>/dev/null
-mkdir build_${BUILD_SUFFIX} && cd build_${BUILD_SUFFIX}
+rm -rf build_${BUILD_SUFFIX}_$1 2>/dev/null
+mkdir build_${BUILD_SUFFIX}_$1 && cd build_${BUILD_SUFFIX}_$1
 
 module load cmake/3.14.5
 
+if [ "$1" == "seq" ]; then
+    argS="On"
+    argV="Off"
+    RAJA_HOSTCONFIG=../tpl/RAJA/host-configs/lc-builds/toss3/clang_X.cmake
+else
+    argS="On"
+    argV="On"
+    RAJA_HOSTCONFIG=../tpl/RAJAvec/host-configs/lc-builds/toss3/clang_X.cmake
+fi
+
+
 cmake \
   -DCMAKE_BUILD_TYPE=Release \
   -DCMAKE_CXX_COMPILER=/usr/tce/packages/clang/clang-10.0.1/bin/clang++ \
+  -DCMAKE_CXX_FLAGS=-ffp-contract=fast \
   -C ${RAJA_HOSTCONFIG} \
   -DENABLE_OPENMP=On \
+  -DENABLE_RAJA_VECTORIZATION=$argV \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   .. 
diff --git a/scripts/lc-builds/toss3_clang9.0.0.sh b/scripts/lc-builds/toss3_clang9.0.0.sh
index 88678d230..5e0f3e195 100755
--- a/scripts/lc-builds/toss3_clang9.0.0.sh
+++ b/scripts/lc-builds/toss3_clang9.0.0.sh
@@ -11,16 +11,27 @@
 BUILD_SUFFIX=lc_toss3-clang-9.0.0
 RAJA_HOSTCONFIG=../tpl/RAJA/host-configs/lc-builds/toss3/clang_X.cmake
 
-rm -rf build_${BUILD_SUFFIX} 2>/dev/null
-mkdir build_${BUILD_SUFFIX} && cd build_${BUILD_SUFFIX}
+rm -rf build_${BUILD_SUFFIX}_$1 2>/dev/null
+mkdir build_${BUILD_SUFFIX}_$1 && cd build_${BUILD_SUFFIX}_$1
 
 module load cmake/3.14.5
 
+if [ "$1" == "seq" ]; then
+    argS="On"
+    argV="Off"
+    RAJA_HOSTCONFIG=../tpl/RAJA/host-configs/lc-builds/toss3/clang_X.cmake
+else
+    argS="On"
+    argV="On"
+    RAJA_HOSTCONFIG=../tpl/RAJAvec/host-configs/lc-builds/toss3/clang_X.cmake
+fi
+
 cmake \
   -DCMAKE_BUILD_TYPE=Release \
   -DCMAKE_CXX_COMPILER=/usr/tce/packages/clang/clang-9.0.0/bin/clang++ \
   -C ${RAJA_HOSTCONFIG} \
   -DENABLE_OPENMP=On \
+  -DENABLE_RAJA_VECTORIZATION=$argV \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   .. 
diff --git a/scripts/lc-builds/toss3_gcc8.1.0.sh b/scripts/lc-builds/toss3_gcc8.1.0.sh
index 4d52cc356..63feb0ca1 100755
--- a/scripts/lc-builds/toss3_gcc8.1.0.sh
+++ b/scripts/lc-builds/toss3_gcc8.1.0.sh
@@ -11,16 +11,27 @@
 BUILD_SUFFIX=lc_toss3-gcc-8.1.0
 RAJA_HOSTCONFIG=../tpl/RAJA/host-configs/lc-builds/toss3/gcc_X.cmake
 
-rm -rf build_${BUILD_SUFFIX} 2>/dev/null
-mkdir build_${BUILD_SUFFIX} && cd build_${BUILD_SUFFIX}
+rm -rf build_${BUILD_SUFFIX}_$1 2>/dev/null
+mkdir build_${BUILD_SUFFIX}_$1 && cd build_${BUILD_SUFFIX}_$1
 
 module load cmake/3.14.5
 
+if [ "$1" == "seq" ]; then
+    argS="On"
+    argV="Off"
+    RAJA_HOSTCONFIG=../tpl/RAJA/host-configs/lc-builds/toss3/clang_X.cmake
+else
+    argS="On"
+    argV="On"
+    RAJA_HOSTCONFIG=../tpl/RAJAvec/host-configs/lc-builds/toss3/clang_X.cmake
+fi
+
 cmake \
   -DCMAKE_BUILD_TYPE=Release \
   -DCMAKE_CXX_COMPILER=/usr/tce/packages/gcc/gcc-8.1.0/bin/g++ \
   -C ${RAJA_HOSTCONFIG} \
   -DENABLE_OPENMP=On \
+  -DENABLE_RAJA_VECTORIZATION=$argV \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss3_icpc19.1.0.sh b/scripts/lc-builds/toss3_icpc19.1.0.sh
index 00df84d16..030b9253a 100755
--- a/scripts/lc-builds/toss3_icpc19.1.0.sh
+++ b/scripts/lc-builds/toss3_icpc19.1.0.sh
@@ -11,16 +11,28 @@
 BUILD_SUFFIX=lc_toss3-icpc-19.1.0
 RAJA_HOSTCONFIG=../tpl/RAJA/host-configs/lc-builds/toss3/icpc_X_gcc8headers.cmake
 
-rm -rf build_${BUILD_SUFFIX} 2>/dev/null
-mkdir build_${BUILD_SUFFIX} && cd build_${BUILD_SUFFIX}
+rm -rf build_${BUILD_SUFFIX}_$1 2>/dev/null
+mkdir build_${BUILD_SUFFIX}_$1 && cd build_${BUILD_SUFFIX}_$1
 
 module load cmake/3.14.5
 
+if [ "$1" == "seq" ]; then
+    argS="On"
+    argV="Off"
+    RAJA_HOSTCONFIG=../tpl/RAJA/host-configs/lc-builds/toss3/clang_X.cmake
+else
+    argS="On"
+    argV="On"
+    RAJA_HOSTCONFIG=../tpl/RAJAvec/host-configs/lc-builds/toss3/clang_X.cmake
+fi
+
 cmake \
   -DCMAKE_BUILD_TYPE=Release \
   -DCMAKE_CXX_COMPILER=/usr/tce/packages/intel/intel-19.1.0/bin/icpc \
+  -DCMAKE_CXX_FLAGS="-xCORE-AVX2" \
   -C ${RAJA_HOSTCONFIG} \
   -DENABLE_OPENMP=On \
+  -DENABLE_RAJA_VECTORIZATION=$argV \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/src/basic/DAXPY-Seq.cpp b/src/basic/DAXPY-Seq.cpp
index 4c792a28f..0b795dbeb 100644
--- a/src/basic/DAXPY-Seq.cpp
+++ b/src/basic/DAXPY-Seq.cpp
@@ -76,7 +76,50 @@ void DAXPY::runSeqVariant(VariantID vid)
 
       break;
     }
+
+    case RAJA_Vec : {
+
+#if(0)
+      DAXPY_DATA_VEC_SETUP;
+
+      auto daxpy_vec_lam = [=](RAJA::VectorIndex<I, vector_t> i) {
+                       DAXPY_VEC_BODY;
+                    };
+      startTimer();
+      for (RepIndex_type irep = 0; irep < run_reps; ++irep) {
+        RAJA::forall<RAJA::vector_exec<vector_t>>(
+          RAJA::TypedRangeSegment<I>(ibegin, iend), daxpy_vec_lam);
+      }
+      stopTimer();
+#endif
+
+#if(0)
+      DAXPY_DATA_VEC_SETUP2;
+
+      auto daxpy_vec_lam = [=](RAJA::VectorIndex<I, vector_t> i) {
+                       DAXPY_VEC_BODY2;
+                    };
+
+      startTimer();
+      for (RepIndex_type irep = 0; irep < run_reps; ++irep) {
+        RAJA::forall<RAJA::vector_exec<vector_t>>(
+          RAJA::TypedRangeSegment<I>(ibegin, iend), daxpy_vec_lam);
+      }
+      stopTimer();
+#endif
+
+#if(1)
+      DAXPY_DATA_VEC_SETUP3;
+
+      startTimer();
+      for (RepIndex_type irep = 0; irep < run_reps; ++irep) {
+        DAXPY_VEC_BODY3;
+      }
+      stopTimer();
 #endif
+      break;
+    }
+#endif //RUN_RAJA_VEC
 
     default : {
       std::cout << "\n  DAXPY : Unknown variant id = " << vid << std::endl;
diff --git a/src/basic/DAXPY.cpp b/src/basic/DAXPY.cpp
index 3edd8da78..5b5788b08 100644
--- a/src/basic/DAXPY.cpp
+++ b/src/basic/DAXPY.cpp
@@ -27,6 +27,7 @@ DAXPY::DAXPY(const RunParams& params)
   setVariantDefined( Base_Seq );
   setVariantDefined( Lambda_Seq );
   setVariantDefined( RAJA_Seq );
+  setVariantDefined( RAJA_Vec );
 
   setVariantDefined( Base_OpenMP );
   setVariantDefined( Lambda_OpenMP );
diff --git a/src/basic/DAXPY.hpp b/src/basic/DAXPY.hpp
index 1d1a56c96..6b822a0a5 100644
--- a/src/basic/DAXPY.hpp
+++ b/src/basic/DAXPY.hpp
@@ -20,11 +20,53 @@
 #define DAXPY_DATA_SETUP \
   Real_ptr x = m_x; \
   Real_ptr y = m_y; \
-  Real_type a = m_a;
+  const Real_type a = m_a;
 
 #define DAXPY_BODY  \
   y[i] += a * x[i] ;
 
+#define DAXPY_DATA_VEC_SETUP \
+  RAJA_INDEX_VALUE_T(I, Int_type, "I");\
+  using vector_t = RAJA::StreamVector<Real_type, 2>; \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> X(x, iend); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> Y(y, iend);
+
+#define DAXPY_DATA_VEC_SETUP2 \
+  RAJA_INDEX_VALUE_T(I, Int_type, "I"); \
+  using vector_t = RAJA::StreamVector<Real_type,2>; \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> Xview(x, iend); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> Yview(y, iend); \
+  RAJA::forall<RAJA::vector_exec<vector_t>> (RAJA::TypedRangeSegment<I>(ibegin, iend),\
+  [=](RAJA::VectorIndex<I, vector_t> i) { \
+    vector_t X(0), Y(0); \
+    for(int j = 0; j < i.size(); ++j) { \
+      X.set(j, *(x + (**i) + j)); \
+      Y.set(j, *(y + (**i) + j)); \
+    } \
+    Xview(i) = X; \
+    Yview(i) = Y; \
+  });
+
+#define DAXPY_DATA_VEC_SETUP3 \
+  RAJA_INDEX_VALUE_T(I, Int_type, "I");\
+  using element_t = RAJA::StreamVector<Real_type,2>::element_type; \
+  element_t X[iend], Y[iend]; \
+  for(int i = 0; i < iend; ++i) { \
+    X[i] = x[i]; \
+    Y[i] = y[i]; \
+  }
+
+#define DAXPY_VEC_BODY \
+  Y(i) += a * X(i);
+
+#define DAXPY_VEC_BODY2 \
+  Yview(i) += a*Xview(i);
+
+#define DAXPY_VEC_BODY3 \
+  for(int i = 0;i < iend; ++i){ \
+    Y[i] += a * X[i]; \
+    y[i] = Y[i]; \
+  }
 
 #include "common/KernelBase.hpp"
 
diff --git a/src/basic/INIT3-Seq.cpp b/src/basic/INIT3-Seq.cpp
index 1483bc9df..0d427f03c 100644
--- a/src/basic/INIT3-Seq.cpp
+++ b/src/basic/INIT3-Seq.cpp
@@ -76,6 +76,25 @@ void INIT3::runSeqVariant(VariantID vid)
 
       break;
     }
+
+    case RAJA_Vec : {
+      INIT3_VEC_SETUP;
+
+      auto init3_vec_lam = [=](RAJA::VectorIndex<I, vector_t> i) {
+                        INIT3_VEC_BODY;
+                     };
+      startTimer();
+      for (RepIndex_type irep = 0; irep < run_reps; ++irep) {
+
+        RAJA::forall<RAJA::vector_exec<vector_t>>(
+          RAJA::TypedRangeSegment<I>(ibegin, iend), init3_vec_lam);
+
+      }
+      stopTimer();
+
+      break;
+    }
+
 #endif // RUN_RAJA_SEQ
 
     default : {
diff --git a/src/basic/INIT3.cpp b/src/basic/INIT3.cpp
index 1adbe906c..baedc297f 100644
--- a/src/basic/INIT3.cpp
+++ b/src/basic/INIT3.cpp
@@ -27,6 +27,7 @@ INIT3::INIT3(const RunParams& params)
   setVariantDefined( Base_Seq );
   setVariantDefined( Lambda_Seq );
   setVariantDefined( RAJA_Seq );
+  setVariantDefined( RAJA_Vec );
 
   setVariantDefined( Base_OpenMP );
   setVariantDefined( Lambda_OpenMP );
diff --git a/src/basic/INIT3.hpp b/src/basic/INIT3.hpp
index 1a0216344..d9ea2b48f 100644
--- a/src/basic/INIT3.hpp
+++ b/src/basic/INIT3.hpp
@@ -25,9 +25,20 @@
   Real_ptr in1 = m_in1; \
   Real_ptr in2 = m_in2;
 
+#define INIT3_VEC_SETUP \
+  RAJA_INDEX_VALUE_T(I, Int_type, "I"); \
+  using vector_t = RAJA::StreamVector<Real_type,2>; \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> O1(out1, iend); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> O2(out2, iend); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> O3(out3, iend); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> I1(in1, iend); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> I2(in2, iend);
+
 #define INIT3_BODY  \
   out1[i] = out2[i] = out3[i] = - in1[i] - in2[i] ;
 
+#define INIT3_VEC_BODY \
+  O1(i) = O2(i) = O3(i) = -1 * I1(i) - I2(i);
 
 #include "common/KernelBase.hpp"
 
diff --git a/src/basic/INIT_VIEW1D-Seq.cpp b/src/basic/INIT_VIEW1D-Seq.cpp
index 89f186b9b..5289f6a76 100644
--- a/src/basic/INIT_VIEW1D-Seq.cpp
+++ b/src/basic/INIT_VIEW1D-Seq.cpp
@@ -82,6 +82,19 @@ void INIT_VIEW1D::runSeqVariant(VariantID vid)
 
       break;
     }
+
+    case RAJA_Vec : {
+
+      INIT_VIEW1D_DATA_VEC_SETUP;
+
+      startTimer();
+      for (RepIndex_type irep = 0; irep < run_reps; ++irep) {
+        INIT_VIEW1D_VEC_BODY;
+      }
+      stopTimer();
+
+      break;
+    }
 #endif // RUN_RAJA_SEQ
 
     default : {
diff --git a/src/basic/INIT_VIEW1D.cpp b/src/basic/INIT_VIEW1D.cpp
index 8f8fed084..a2e220e7c 100644
--- a/src/basic/INIT_VIEW1D.cpp
+++ b/src/basic/INIT_VIEW1D.cpp
@@ -27,6 +27,7 @@ INIT_VIEW1D::INIT_VIEW1D(const RunParams& params)
   setVariantDefined( Base_Seq );
   setVariantDefined( Lambda_Seq );
   setVariantDefined( RAJA_Seq );
+  setVariantDefined( RAJA_Vec );
 
   setVariantDefined( Base_OpenMP );
   setVariantDefined( Lambda_OpenMP );
diff --git a/src/basic/INIT_VIEW1D.hpp b/src/basic/INIT_VIEW1D.hpp
index 77a9eebe6..cb0ecf463 100644
--- a/src/basic/INIT_VIEW1D.hpp
+++ b/src/basic/INIT_VIEW1D.hpp
@@ -39,6 +39,20 @@
   const RAJA::Layout<1> my_layout(iend); \
   ViewType view(a, my_layout);
 
+#define INIT_VIEW1D_DATA_VEC_SETUP \
+  RAJA_INDEX_VALUE_T(I, Int_type, "I"); \
+  using vector_t = RAJA::StreamVector<Real_type, 2>; \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> Aview(a, iend);
+
+#define INIT_VIEW1D_VEC_BODY  \
+  RAJA::forall<RAJA::vector_exec<vector_t>> (RAJA::TypedRangeSegment<I>(ibegin, iend),\
+  [=](RAJA::VectorIndex<I, vector_t> i) { \
+    vector_t A(0); \
+    for(int j = 0; j < i.size(); ++j) { \
+      A.set(j, (**i + j + 1) * v); \
+    } \
+    Aview(i) = A; \
+  });
 
 #include "common/KernelBase.hpp"
 
diff --git a/src/basic/MULADDSUB-Seq.cpp b/src/basic/MULADDSUB-Seq.cpp
index dfda9d17b..62a6c9622 100644
--- a/src/basic/MULADDSUB-Seq.cpp
+++ b/src/basic/MULADDSUB-Seq.cpp
@@ -76,6 +76,27 @@ void MULADDSUB::runSeqVariant(VariantID vid)
 
       break;
     }
+
+    case RAJA_Vec : {
+
+      MULADDSUB_DATA_VEC_SETUP;
+
+      auto mas_lam = [=](RAJA::VectorIndex<I, vector_t> i) {
+                       MULADDSUB_VEC_BODY;
+                     };
+
+      startTimer();
+      for (RepIndex_type irep = 0; irep < run_reps; ++irep) {
+
+        RAJA::forall<RAJA::vector_exec<vector_t>>(
+          RAJA::TypedRangeSegment<I>(ibegin, iend), mas_lam);
+
+      }
+      stopTimer();
+
+      break;
+    }
+
 #endif // RUN_RAJA_SEQ
 
     default : {
diff --git a/src/basic/MULADDSUB.cpp b/src/basic/MULADDSUB.cpp
index fd46c3718..7568ad3ea 100644
--- a/src/basic/MULADDSUB.cpp
+++ b/src/basic/MULADDSUB.cpp
@@ -27,6 +27,7 @@ MULADDSUB::MULADDSUB(const RunParams& params)
   setVariantDefined( Base_Seq );
   setVariantDefined( Lambda_Seq );
   setVariantDefined( RAJA_Seq );
+  setVariantDefined( RAJA_Vec );
 
   setVariantDefined( Base_OpenMP );
   setVariantDefined( Lambda_OpenMP );
diff --git a/src/basic/MULADDSUB.hpp b/src/basic/MULADDSUB.hpp
index 48a664f63..9a9a7b376 100644
--- a/src/basic/MULADDSUB.hpp
+++ b/src/basic/MULADDSUB.hpp
@@ -31,6 +31,19 @@
   out2[i] = in1[i] + in2[i] ; \
   out3[i] = in1[i] - in2[i] ;
 
+#define MULADDSUB_DATA_VEC_SETUP \
+  RAJA_INDEX_VALUE_T(I, Int_type, "I"); \
+  using vector_t = RAJA::StreamVector<Real_type, 2>; \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> O1(out1, iend); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> O2(out2, iend); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> O3(out3, iend); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> I1(in1, iend); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> I2(in2, iend); 
+
+#define MULADDSUB_VEC_BODY \
+  O1(i) = I1(i) * I2(i); \
+  O2(i) = I1(i) + I2(i); \
+  O3(i) = I1(i) - I2(i);
 
 #include "common/KernelBase.hpp"
 
diff --git a/src/common/KernelBase.cpp b/src/common/KernelBase.cpp
index f339cbe3a..ae34b5a56 100644
--- a/src/common/KernelBase.cpp
+++ b/src/common/KernelBase.cpp
@@ -97,6 +97,7 @@ void KernelBase::runKernel(VariantID vid)
     case Base_Seq :
     case Lambda_Seq :
     case RAJA_Seq :
+    case RAJA_Vec :
     {
 #if defined(RUN_RAJA_SEQ)
       runSeqVariant(vid);
diff --git a/src/common/RAJAPerfSuite.cpp b/src/common/RAJAPerfSuite.cpp
index a5581e196..6c01e7c1c 100644
--- a/src/common/RAJAPerfSuite.cpp
+++ b/src/common/RAJAPerfSuite.cpp
@@ -213,6 +213,7 @@ static const std::string VariantNames [] =
   std::string("Base_Seq"),
   std::string("Lambda_Seq"),
   std::string("RAJA_Seq"),
+  std::string("RAJA_Vec"),
 
   std::string("Base_OpenMP"),
   std::string("Lambda_OpenMP"),
@@ -302,7 +303,8 @@ bool isVariantAvailable(VariantID vid)
   }
 #if defined(RUN_RAJA_SEQ)
   if ( vid == Lambda_Seq || 
-       vid == RAJA_Seq ) {
+       vid == RAJA_Seq ||
+       vid == RAJA_Vec) {
     ret_val = true;
   }
 #endif
diff --git a/src/common/RAJAPerfSuite.hpp b/src/common/RAJAPerfSuite.hpp
index 30675df3e..d6f964126 100644
--- a/src/common/RAJAPerfSuite.hpp
+++ b/src/common/RAJAPerfSuite.hpp
@@ -184,6 +184,7 @@ enum VariantID {
   Base_Seq = 0,
   Lambda_Seq,
   RAJA_Seq,
+  RAJA_Vec,
 
   Base_OpenMP,
   Lambda_OpenMP,
diff --git a/src/lcals/FIRST_SUM-Seq.cpp b/src/lcals/FIRST_SUM-Seq.cpp
index bc05f7e5f..e45f00516 100644
--- a/src/lcals/FIRST_SUM-Seq.cpp
+++ b/src/lcals/FIRST_SUM-Seq.cpp
@@ -76,6 +76,19 @@ void FIRST_SUM::runSeqVariant(VariantID vid)
 
       break;
     }
+
+    case RAJA_Vec : {
+
+      FIRST_SUM_VEC_SETUP;
+
+      startTimer();
+      for (RepIndex_type irep = 0; irep < run_reps; ++irep) {
+        FIRST_SUM_VEC_BODY;
+      }
+      stopTimer();
+
+      break;
+    }
 #endif // RUN_RAJA_SEQ
 
     default : {
diff --git a/src/lcals/FIRST_SUM.cpp b/src/lcals/FIRST_SUM.cpp
index c849efdf7..f98793414 100644
--- a/src/lcals/FIRST_SUM.cpp
+++ b/src/lcals/FIRST_SUM.cpp
@@ -27,6 +27,7 @@ FIRST_SUM::FIRST_SUM(const RunParams& params)
   setVariantDefined( Base_Seq );
   setVariantDefined( Lambda_Seq );
   setVariantDefined( RAJA_Seq );
+  setVariantDefined( RAJA_Vec );
                      
   setVariantDefined( Base_OpenMP );
   setVariantDefined( Lambda_OpenMP );
diff --git a/src/lcals/FIRST_SUM.hpp b/src/lcals/FIRST_SUM.hpp
index 76b2c3552..14a9e78ec 100644
--- a/src/lcals/FIRST_SUM.hpp
+++ b/src/lcals/FIRST_SUM.hpp
@@ -25,6 +25,21 @@
   Real_ptr x = m_x; \
   Real_ptr y = m_y;
 
+#define FIRST_SUM_VEC_SETUP \
+  RAJA_INDEX_VALUE_T(I, Int_type, "I"); \
+  using element_t = RAJA::StreamVector<Real_type,2>::element_type; \
+  element_t X[iend], Y[iend]; \
+  for(int i = 0; i < iend; ++i) { \
+    X[i] = x[i]; \
+    Y[i] = y[i]; \
+  }
+
+#define FIRST_SUM_VEC_BODY \
+  for(int i = 1; i < iend; ++i) { \
+    X[i] = Y[i-1] + Y[i]; \
+    x[i] = X[i]; \
+  }
+
 #define FIRST_SUM_BODY  \
   x[i] = y[i-1] + y[i];
 
diff --git a/src/polybench/POLYBENCH_2MM-Seq.cpp b/src/polybench/POLYBENCH_2MM-Seq.cpp
index 4485ef2d8..64fe36800 100644
--- a/src/polybench/POLYBENCH_2MM-Seq.cpp
+++ b/src/polybench/POLYBENCH_2MM-Seq.cpp
@@ -156,10 +156,10 @@ void POLYBENCH_2MM::runSeqVariant(VariantID vid)
                            RAJA::RangeSegment{0, nk}),
           RAJA::tuple<Real_type>{0.0},
 
-          poly_2mm_lam1, 
-          poly_2mm_lam2, 
+          poly_2mm_lam1,
+          poly_2mm_lam2,
           poly_2mm_lam3
-        );
+	);
 
         RAJA::kernel_param<EXEC_POL>( 
           RAJA::make_tuple(RAJA::RangeSegment{0, ni},
@@ -177,6 +177,19 @@ void POLYBENCH_2MM::runSeqVariant(VariantID vid)
       break;
     }
 
+    case RAJA_Vec : {
+
+      POLYBENCH_2MM_DATA_VEC_SETUP; 
+
+      startTimer();
+      for (RepIndex_type irep = 0; irep < run_reps; ++irep) {
+        POLYBENCH_2MM_VEC_BODY1;
+      }
+      stopTimer();
+
+      break;
+
+    }
 #endif // RUN_RAJA_SEQ
 
     default : {
diff --git a/src/polybench/POLYBENCH_2MM.cpp b/src/polybench/POLYBENCH_2MM.cpp
index 3bac5f0ad..a81d4dcc7 100644
--- a/src/polybench/POLYBENCH_2MM.cpp
+++ b/src/polybench/POLYBENCH_2MM.cpp
@@ -58,6 +58,7 @@ POLYBENCH_2MM::POLYBENCH_2MM(const RunParams& params)
   setVariantDefined( Base_Seq );
   setVariantDefined( Lambda_Seq );
   setVariantDefined( RAJA_Seq );
+  setVariantDefined( RAJA_Vec );
                      
   setVariantDefined( Base_OpenMP );
   setVariantDefined( Lambda_OpenMP );
diff --git a/src/polybench/POLYBENCH_2MM.hpp b/src/polybench/POLYBENCH_2MM.hpp
index 1171fb185..22c77f0eb 100644
--- a/src/polybench/POLYBENCH_2MM.hpp
+++ b/src/polybench/POLYBENCH_2MM.hpp
@@ -55,7 +55,7 @@
   Real_type dot = 0.0;
 
 #define POLYBENCH_2MM_BODY2 \
-  dot += alpha * A[k + i*nk] * B[j + k*nj];
+  dot += alpha * A[k + i*nk] * B[j + k*nj];\
 
 #define POLYBENCH_2MM_BODY3 \
   tmp[j + i*nj] = dot;
@@ -99,6 +99,97 @@ using VIEW_TYPE = RAJA::View<Real_type, \
   VIEW_TYPE Cview(C, RAJA::Layout<2>(nj, nl)); \
   VIEW_TYPE Dview(D, RAJA::Layout<2>(ni, nl));
 
+#define POLYBENCH_2MM_DATA_VEC_SETUP \
+  RAJA_INDEX_VALUE_T(II, Int_type, "II"); \
+  RAJA_INDEX_VALUE_T(IJ, Int_type, "IJ"); \
+  RAJA_INDEX_VALUE_T(IK, Int_type, "IK"); \
+  RAJA_INDEX_VALUE_T(IL, Int_type, "IL"); \
+  using matrix_t = RAJA::RegisterMatrix<Real_type, RAJA::MATRIX_ROW_MAJOR>; \
+  std::array<RAJA::idx_t, 2> perm {{0,1}}; \
+  RAJA::TypedView<Real_type, RAJA::Layout<2, Int_type, 1>, II, IK> Aview(A, RAJA::make_permuted_layout({{ni, nk}}, perm)); \
+  RAJA::TypedView<Real_type, RAJA::Layout<2, Int_type, 1>, IK, IJ> Bview(B, RAJA::make_permuted_layout({{nk, nj}}, perm)); \
+  RAJA::TypedView<Real_type, RAJA::Layout<2, Int_type, 1>, II, IJ> Tmpview(tmp, RAJA::make_permuted_layout({{ni, nj}}, perm)); \
+  RAJA::TypedView<Real_type, RAJA::Layout<2, Int_type, 1>, IJ, IL> Cview(C, RAJA::make_permuted_layout({{nj, nl}}, perm)); \
+  RAJA::TypedView<Real_type, RAJA::Layout<2, Int_type, 1>, II, IL> Dview(D, RAJA::make_permuted_layout({{ni, nl}}, perm)); \
+  using RowA = RAJA::RowIndex<II, matrix_t>; \
+  using ColA = RAJA::ColIndex<IK, matrix_t>; \
+  using ColB = RAJA::ColIndex<IJ, matrix_t>; \
+  using RowT = RAJA::RowIndex<II, matrix_t>; \
+  using ColT = RAJA::ColIndex<IJ, matrix_t>; \
+  using ColC = RAJA::ColIndex<IL, matrix_t>; \
+  using EXECPOL = \
+    RAJA::KernelPolicy< \
+      RAJA::statement::For<2, RAJA::matrix_col_exec<matrix_t>, \
+        RAJA::statement::For<1, RAJA::matrix_col_exec<matrix_t>, \
+          RAJA::statement::For<0, RAJA::matrix_row_exec<matrix_t>, \
+            RAJA::statement::Lambda<0> \
+            > \
+          > \
+        > \
+      >; \
+  for(int i = 0; i < ni*nk; i++) { \
+    A[i] = A[i] * alpha; \
+  }
+
+#define POLYBENCH_2MM_VEC_BODY1 \
+  std::memset(tmp, 0, ni*nj* sizeof(Real_type));\
+  for(int i = 0; i < ni*nl; i++) { \
+    D[i] = beta; \
+  } \
+  auto segments1 = RAJA::make_tuple(RAJA::TypedRangeSegment<II>(0, ni),\
+                                   RAJA::TypedRangeSegment<IK>(0, nk),\
+                                   RAJA::TypedRangeSegment<IJ>(0, nj)); \
+  RAJA::kernel<EXECPOL>( segments1, \
+      [=] (RowA i, ColA k, ColB j) { \
+          Tmpview(i, j) += Aview(i, k) * Bview(toRowIndex(k), j); \
+      } \
+  );\
+  auto segments2 = RAJA::make_tuple(RAJA::TypedRangeSegment<II>(0, ni),\
+                                   RAJA::TypedRangeSegment<IJ>(0, nj),\
+                                   RAJA::TypedRangeSegment<IL>(0, nl)); \
+  RAJA::kernel<EXECPOL>( segments2, \
+      [=] (RowT i, ColT j, ColC l) { \
+          Dview(i, l) += Tmpview(i, j) * Cview(toRowIndex(j), l); \
+      } \
+  ); 
+//  for (II i(0); i < ni; i++ ) { \
+//    for (IL l(0); l < nl; l++) { \
+//      dot = 0; \
+//      for (IJ j(0); j < nj; j++) { \
+//        dot += Tmpview(i,j) * Cview(j,l); \
+//        std::cout << "Tmp and C view: " << Tmpview(i,j) << " " << Cview(j,l) << " " << dot << std::endl; \
+//      } \
+//      std::cout << dot << " " << Dview(i,l) << std::endl;\
+//    } \
+//  }
+//  for (II i(0); i < ni; i++ ) { \
+//    for (IJ j(0); j < nj; j++) { \
+//      dot = 0; \
+//      for (IK k(0); k < nk; k++) { \
+//        dot += Aview(i,k) * Bview(k,j); \
+//        std::cout << "A and B view: " << Aview(i,k) << " " << Bview(k,j) << " " << dot << std::endl; \
+//      } \
+//      std::cout << dot << " " << Tmpview(i,j) << std::endl;\
+//    } \
+//  }
+
+//#define POLYBENCH_2MM_VEC_BODY1 \
+//  for (Index_type i = 0; i < ni; i++ ) { \
+//    for (Index_type j = 0; j < nj; j++) { \
+//      dot = 0.; \
+//      for (Index_type k = 0; k < nk; k++) { \
+//        dot += alpha * a.get(i,k) * b.get(k,j); \
+//        std::cout << "dot: " << dot << " " << a.get(i,k) << " " << a.get(k,i) << " " << A[k+i*nk] << " " << b.get(j,k) << " " << b.get(k,j) << " " << B[j+k*nj] << std::endl; \
+//      } \
+//      c.set(i,j,dot); \
+//    } \
+//  }
+        //std::cout << "dot: " << dot << " " << a.get(i,k) << " " << a.get(k,i) << " " << A[i+k*ni] << " " << b.get(j,k) << " " << b.get(k,j) << " " << B[k+j*nk] << std::endl; \
+  //auto c = a * b; \
+  //for (Index_type i = 0; i < ni; i++ ) { \
+  //  for (Index_type j = 0; j < nj; j++) { \
+  //  } \
+  //}
 
 #include "common/KernelBase.hpp"
 
diff --git a/src/polybench/POLYBENCH_GEMVER-OMP.cpp b/src/polybench/POLYBENCH_GEMVER-OMP.cpp
index 843ad1608..60b3ef761 100644
--- a/src/polybench/POLYBENCH_GEMVER-OMP.cpp
+++ b/src/polybench/POLYBENCH_GEMVER-OMP.cpp
@@ -198,8 +198,8 @@ void POLYBENCH_GEMVER::runOpenMPVariant(VariantID vid)
       startTimer();
       for (RepIndex_type irep = 0; irep < run_reps; ++irep) {
 
-        RAJA::kernel<EXEC_POL1>( RAJA::make_tuple(RAJA::RangeSegment{0, n},
-                                                  RAJA::RangeSegment{0, n}),
+        RAJA::kernel<EXEC_POL1>( RAJA::make_tuple(RAJA::RangeSegment(0, n),
+                                                  RAJA::RangeSegment(0, n)),
           poly_gemver_lam1
         );
 
diff --git a/src/polybench/POLYBENCH_GEMVER-Seq.cpp b/src/polybench/POLYBENCH_GEMVER-Seq.cpp
index a1563e84b..46d15ef5f 100644
--- a/src/polybench/POLYBENCH_GEMVER-Seq.cpp
+++ b/src/polybench/POLYBENCH_GEMVER-Seq.cpp
@@ -189,8 +189,8 @@ void POLYBENCH_GEMVER::runSeqVariant(VariantID vid)
       startTimer();
       for (RepIndex_type irep = 0; irep < run_reps; ++irep) {
 
-        RAJA::kernel<EXEC_POL1>( RAJA::make_tuple(RAJA::RangeSegment{0, n},
-                                                  RAJA::RangeSegment{0, n}),
+        RAJA::kernel<EXEC_POL1>( RAJA::make_tuple(RAJA::RangeSegment(0, n),
+                                                  RAJA::RangeSegment(0, n)),
           poly_gemver_lam1
         );
         
@@ -204,7 +204,7 @@ void POLYBENCH_GEMVER::runSeqVariant(VariantID vid)
           poly_gemver_lam4
         );
         
-        RAJA::forall<EXEC_POL3> (RAJA::RangeSegment{0, n}, 
+        RAJA::forall<EXEC_POL3> (RAJA::RangeSegment(0, n), 
           poly_gemver_lam5
         );
 
diff --git a/src/stream/ADD-Seq.cpp b/src/stream/ADD-Seq.cpp
index 3ef8b81c9..387746f86 100644
--- a/src/stream/ADD-Seq.cpp
+++ b/src/stream/ADD-Seq.cpp
@@ -76,7 +76,27 @@ void ADD::runSeqVariant(VariantID vid)
 
       break;
     }
-#endif // RUN_RAJA_SEQ
+
+    case RAJA_Vec : {
+
+      ADD_DATA_VEC_SETUP;
+
+      auto add_vec_lam = [=](RAJA::VectorIndex<I, vector_t> i) {
+                       ADD_VEC_BODY;
+                 };
+
+      startTimer();
+      for (RepIndex_type irep = 0; irep < run_reps; ++irep) {
+        
+        RAJA::forall<RAJA::vector_exec<vector_t>>(
+          RAJA::TypedRangeSegment<I>(ibegin, iend), add_vec_lam);
+
+      }
+      stopTimer();
+
+      break;
+    }
+#endif //RUN_RAJA_SEQ
 
     default : {
       std::cout << "\n  ADD : Unknown variant id = " << vid << std::endl;
diff --git a/src/stream/ADD.cpp b/src/stream/ADD.cpp
index 00cb4ecf0..76287c8d7 100644
--- a/src/stream/ADD.cpp
+++ b/src/stream/ADD.cpp
@@ -27,6 +27,7 @@ ADD::ADD(const RunParams& params)
   setVariantDefined( Base_Seq );
   setVariantDefined( Lambda_Seq );
   setVariantDefined( RAJA_Seq );
+  setVariantDefined( RAJA_Vec );
                      
   setVariantDefined( Base_OpenMP );
   setVariantDefined( Lambda_OpenMP );
diff --git a/src/stream/ADD.hpp b/src/stream/ADD.hpp
index 542c253b0..f54ccc981 100644
--- a/src/stream/ADD.hpp
+++ b/src/stream/ADD.hpp
@@ -20,11 +20,27 @@
 #define ADD_DATA_SETUP \
   Real_ptr a = m_a; \
   Real_ptr b = m_b; \
-  Real_ptr c = m_c;
+  Real_ptr c = m_c; \
+  RAJA_INDEX_VALUE_T(I, int, "I");\
+  using vector_t = RAJA::StreamVector<Real_type, 16>;\
+  using VecI = RAJA::VectorIndex<I, vector_t>;\
+  RAJA::TypedView<Real_type, RAJA::Layout<1>, I> A(a, getRunSize()); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1>, I> B(b, getRunSize()); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1>, I> C(c, getRunSize());
+
 
 #define ADD_BODY  \
   c[i] = a[i] + b[i]; 
 
+#define ADD_DATA_VEC_SETUP \
+  RAJA_INDEX_VALUE_T(I, Int_type, "I"); \
+  using vector_t = RAJA::StreamVector<Real_type, 2>; \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> A(a, iend); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> B(b, iend); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> C(c, iend); 
+
+#define ADD_VEC_BODY \
+ C(i) = A(i) + B(i);
 
 #include "common/KernelBase.hpp"
 
diff --git a/src/stream/COPY-Seq.cpp b/src/stream/COPY-Seq.cpp
index 68bc51e4d..45e218418 100644
--- a/src/stream/COPY-Seq.cpp
+++ b/src/stream/COPY-Seq.cpp
@@ -76,6 +76,26 @@ void COPY::runSeqVariant(VariantID vid)
 
       break;
     }
+    case RAJA_Vec : {
+
+      COPY_DATA_VEC_SETUP;
+
+      auto copy_vec_lam = [=](RAJA::VectorIndex<I, vector_t> i) {
+                       COPY_VEC_BODY;
+                 };
+
+      startTimer();
+      for (RepIndex_type irep = 0; irep < run_reps; ++irep) {
+
+        RAJA::forall<RAJA::vector_exec<vector_t>>(
+          RAJA::TypedRangeSegment<I>(ibegin, iend), copy_vec_lam);
+
+      }
+      stopTimer();
+
+      break;
+    }
+
 #endif // RUN_RAJA_SEQ
 
     default : {
diff --git a/src/stream/COPY.cpp b/src/stream/COPY.cpp
index 8da987150..761e2e15b 100644
--- a/src/stream/COPY.cpp
+++ b/src/stream/COPY.cpp
@@ -27,6 +27,7 @@ COPY::COPY(const RunParams& params)
   setVariantDefined( Base_Seq );
   setVariantDefined( Lambda_Seq );
   setVariantDefined( RAJA_Seq );
+  setVariantDefined( RAJA_Vec );
                      
   setVariantDefined( Base_OpenMP );
   setVariantDefined( Lambda_OpenMP );
diff --git a/src/stream/COPY.hpp b/src/stream/COPY.hpp
index c0e2c455a..566e962b0 100644
--- a/src/stream/COPY.hpp
+++ b/src/stream/COPY.hpp
@@ -21,9 +21,17 @@
   Real_ptr a = m_a; \
   Real_ptr c = m_c;
 
+#define COPY_DATA_VEC_SETUP \
+  RAJA_INDEX_VALUE_T(I, Int_type, "I"); \
+  using vector_t = RAJA::StreamVector<Real_type, 2>; \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> A(a, iend); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type, 0>, I> C(c, iend); 
+
 #define COPY_BODY  \
   c[i] = a[i] ;
 
+#define COPY_VEC_BODY \
+ C(i) = A(i);
 
 #include "common/KernelBase.hpp"
 
diff --git a/src/stream/MUL-Seq.cpp b/src/stream/MUL-Seq.cpp
index 7b36935e1..53bc6b8fa 100644
--- a/src/stream/MUL-Seq.cpp
+++ b/src/stream/MUL-Seq.cpp
@@ -76,6 +76,26 @@ void MUL::runSeqVariant(VariantID vid)
 
       break;
     }
+
+    case RAJA_Vec : {
+
+      MUL_DATA_VEC_SETUP;
+
+      auto mul_vec_lam = [=](RAJA::VectorIndex<I, vector_t> i) {
+                       MUL_VEC_BODY;
+                 };
+
+      startTimer();
+      for (RepIndex_type irep = 0; irep < run_reps; ++irep) {
+
+        RAJA::forall<RAJA::vector_exec<vector_t>>(
+          RAJA::TypedRangeSegment<I>(ibegin, iend), mul_vec_lam);
+
+      }
+      stopTimer();
+
+      break;
+    }
 #endif // RUN_RAJA_SEQ
 
     default : {
diff --git a/src/stream/MUL.cpp b/src/stream/MUL.cpp
index 0dc16b674..8b17d8962 100644
--- a/src/stream/MUL.cpp
+++ b/src/stream/MUL.cpp
@@ -27,6 +27,7 @@ MUL::MUL(const RunParams& params)
   setVariantDefined( Base_Seq );
   setVariantDefined( Lambda_Seq );
   setVariantDefined( RAJA_Seq );
+  setVariantDefined( RAJA_Vec );
                      
   setVariantDefined( Base_OpenMP );
   setVariantDefined( Lambda_OpenMP );
diff --git a/src/stream/MUL.hpp b/src/stream/MUL.hpp
index 25943fcbe..33e77ed8f 100644
--- a/src/stream/MUL.hpp
+++ b/src/stream/MUL.hpp
@@ -22,9 +22,17 @@
   Real_ptr c = m_c; \
   Real_type alpha = m_alpha;
 
+#define MUL_DATA_VEC_SETUP \
+  RAJA_INDEX_VALUE_T(I, Int_type, "I"); \
+  using vector_t = RAJA::StreamVector<Real_type, 2>; \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type>, I> B(b, iend); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type>, I> C(c, iend); 
+
 #define MUL_BODY  \
   b[i] = alpha * c[i] ;
 
+#define MUL_VEC_BODY \
+ B(i) = alpha * C(i);
 
 #include "common/KernelBase.hpp"
 
diff --git a/src/stream/TRIAD-Seq.cpp b/src/stream/TRIAD-Seq.cpp
index 06885d50c..3829d18f1 100644
--- a/src/stream/TRIAD-Seq.cpp
+++ b/src/stream/TRIAD-Seq.cpp
@@ -76,6 +76,27 @@ void TRIAD::runSeqVariant(VariantID vid)
 
       break;
     }
+
+    case RAJA_Vec : {
+
+      TRIAD_DATA_VEC_SETUP;
+
+      auto triad_vec_lam = [=](RAJA::VectorIndex<I, vector_t> i) {
+                       TRIAD_VEC_BODY;
+                 };
+
+      startTimer();
+      for (RepIndex_type irep = 0; irep < run_reps; ++irep) {
+
+        RAJA::forall<RAJA::vector_exec<vector_t>>(
+          RAJA::TypedRangeSegment<I>(ibegin, iend), triad_vec_lam);
+
+      }
+      stopTimer();
+
+      break;
+    }
+
 #endif // RUN_RAJA_SEQ
 
     default : {
diff --git a/src/stream/TRIAD.cpp b/src/stream/TRIAD.cpp
index 3e42feeda..33eb8b3a3 100644
--- a/src/stream/TRIAD.cpp
+++ b/src/stream/TRIAD.cpp
@@ -27,6 +27,7 @@ TRIAD::TRIAD(const RunParams& params)
   setVariantDefined( Base_Seq );
   setVariantDefined( Lambda_Seq );
   setVariantDefined( RAJA_Seq );
+  setVariantDefined( RAJA_Vec );
                      
   setVariantDefined( Base_OpenMP );
   setVariantDefined( Lambda_OpenMP );
diff --git a/src/stream/TRIAD.hpp b/src/stream/TRIAD.hpp
index 23ffd168a..3952b71f6 100644
--- a/src/stream/TRIAD.hpp
+++ b/src/stream/TRIAD.hpp
@@ -23,9 +23,18 @@
   Real_ptr c = m_c; \
   Real_type alpha = m_alpha;
 
+#define TRIAD_DATA_VEC_SETUP \
+  RAJA_INDEX_VALUE_T(I, Int_type, "I"); \
+  using vector_t = RAJA::StreamVector<Real_type, 2>; \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type>, I> A(a, iend); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type>, I> B(b, iend); \
+  RAJA::TypedView<Real_type, RAJA::Layout<1, Int_type>, I> C(c, iend); 
+
 #define TRIAD_BODY  \
   a[i] = b[i] + alpha * c[i] ;
 
+#define TRIAD_VEC_BODY \
+ A(i) = B(i) + alpha * C(i);
 
 #include "common/KernelBase.hpp"
 
diff --git a/tpl/RAJAvec b/tpl/RAJAvec
new file mode 160000
index 000000000..e3dc2f988
--- /dev/null
+++ b/tpl/RAJAvec
@@ -0,0 +1 @@
+Subproject commit e3dc2f988042f785298719b530867e92192a3c0e